test: keep child deadlines that rule out fallback exits

These fixtures' descendants exit on their own after 20 or 30 seconds. The
shorter explicit deadline is what proves cleanup stopped them first.

Co-Authored-By: gensokyo-dev <[email protected]>
Co-Authored-By: Claude Opus 5.5 <[email protected]>

.agents/skills/reverse-engineer-anything/SKILL.md

@@ -2,123 +2,607 @@
 name: reverse-engineer-anything
 description: Reverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.
 metadata:
-  version: "64"
+  version: "62"
 ---
 
 # REA
 
-Use REA when a claim depends on a shipped artifact, decompilation, requested
-runtime evidence, or behavior not established by available source. For ordinary
-analysis of a complete source repository, use normal repository tools and do not run REA
+Use REA when a claim depends on a shipped binary or package, decompilation,
+requested runtime evidence, or comparison with
+behavior not established by available source. For ordinary analysis of a
+complete source repository, use normal repository tools and do not run REA
 readiness or provider commands.
 
-## Select the target and authority
-
-Use the target selected for this task. If the user says only "my app", ask which
-app before searching artifacts. Resolve a named app to one clear installed
-artifact; ask when ambiguous. Unrelated reports and nearby fixtures do not select
-it. Never substitute an example or broaden the selected target or runtime actions.
-
-Distinguish observations, derivations, inferences, and unknowns. Cite Evidence
-IDs and preserve source identity, coverage, limitations and partial facts.
-Static code, strings and declarations do not establish observed execution.
-Within the requested investigation, call available analysis tools directly;
-REA requires no permission grants or per-call approval flags. The client may
-independently require approval; preserve its actual access requirements.
-
 ## Connect only when needed
 
-If REA tools are available and registration is not known stale, proceed directly.
-Use the connected server's actual tool list and schemas: repository-main guidance
-may describe features absent from an older release. Keep complete inline Evidence
-when retained references are not advertised. Skill installation alone adds no
-MCP server or engines; a Claude Code, Codex or Cursor Marketplace plugin bundles its own server.
+Installing this skill supplies instructions; it does not register the REA MCP
+server, install analysis engines, or add tools to an already-running agent.
+The exception is a Claude Code, Codex, or Cursor Marketplace plugin install:
+the plugin bundles this skill and registers its own `rea` server, outside the
+state that `rea setup` and `rea doctor` manage.
 
-For absent tools, blocked calls, unavailable providers, host timeouts or startup
-failures, read [connection and recovery](references/connection-and-recovery.md)
-before repair. Diagnose the observed blocker, continue useful supported work,
-and resume the original task after repair. Do not repeat unchanged failed calls.
-A host timeout without a REA result does not establish cancellation or cleanup.
-Retry a dropped connection once only if safe to repeat or known not started;
-if it may have changed the target, or fails again, ask for a client restart.
-Retry failed connection startup once; do not retry the same unchanged failure.
+If REA tools are available and their registration is not known to be stale,
+proceed directly to the target. Do not run diagnostics or setup before every
+investigation. Use the connected server's actual tool list and input schemas;
+a skill installed from repository main may describe capabilities absent from
+an older npm release. Keep complete inline Evidence when that server does not
+advertise retained references.
 
-Choose the registration owner: plugin recovery differs from setup-owned repair.
+If tools are absent, a call is blocked by the client, or a required provider is
+unavailable, read [connection and recovery](references/connection-and-recovery.md).
+Diagnose the observed blocker, continue useful supported work, and resume the
+original task after repair. Client approval, shell PATH, registration and engine
+availability have different remedies; avoid repeating setup for all of them.
+If the host times out without a REA result, do not repeat the unchanged call or
+assume work stopped. Read the recovery guide's deadline distinction before retrying.
+If the MCP connection drops, retry that call once only when it is safe to
+repeat or known not to have started. If it may already have changed the
+target, or it fails again, tell the user and ask them to restart the MCP
+client. If REA could not start its connection, retry once, then tell the user
+and ask them to restart the MCP client. Do not retry a startup failure that
+returns the same unchanged result. For plugin installs, follow
+[plugin recovery](references/connection-and-recovery.md#rea-installed-as-a-client-plugin).
+For setup-owned registrations, run `rea doctor` when REA could not start
+or tools stay missing after that restart. For `cleanup_incomplete`, a retryable
+capacity limit, or `configuration_invalid`, follow the recovery guide's ladder
+before asking for a restart. Read that guide before a second repair attempt.
+
+Choose the installation owner before repairing registration. Plugin installs use
+[plugin recovery](references/connection-and-recovery.md#rea-installed-as-a-client-plugin);
+only setup-owned registrations use the guide's
+[registration repair plan](references/connection-and-recovery.md#repair-registration-when-needed).
 A plugin-only doctor result of `missing` does not call for `rea setup`.
-Setup-owned repair starts with a read-only plan showing paths, backups and skill
-replacement before authorized writes. Do not install unrelated software to repair
-registration. REA setup never installs/upgrades Node, npm, Homebrew, Java,
-Ghidra, IDA, JADX, Kuna, Binwalk or Unblob.
-For an unsupported client, use manual stdio with an existing absolute runtime
-and entry script, or the CLI fallback in the recovery guide.
 
-## Route before analysis
+REA setup never installs or upgrades Node.js, npm, Homebrew, Java, Ghidra, IDA,
+JADX, Kuna, Binwalk, or Unblob. Use existing prerequisites; do not install unrelated
+software to repair MCP registration. For an unsupported client, use manual stdio
+registration with an existing installation’s absolute runtime and entry script, or continue through the CLI.
 
-Read the matching guide before its workflow; read only the relevant sections.
-Target-free tools take their explicit paths/endpoints without `open_binary`.
-Active-target native/archive tools require it. Archive members and executable
-code are different questions; follow the selected artifact's application layer.
+A concrete CLI fallback for an operator-supplied JavaScript tree or ASAR is:
 
-| Supplied target or question                                                                                                                                        | Guide and first operation                                                                                                                                                                                                                          |
-| ------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| Native executable/library, Ghidra/IDA/Kuna, managed PE/CLI or NativeAOT                                                                                            | [Native and managed](references/native-and-artifacts.md): `open_binary`; managed CLI metadata: `inspect_managed_artifact`. Read provider/lifecycle sections before reuse or edits.                                                                 |
-| Headerless/raw firmware, Cortex-M, ESP app/bootloader or whole-flash dump, HEX/S-record/UF2/DfuSe                                                                  | [Firmware and raw images](references/firmware-and-raw-images.md): check loader support first; raw: explicit `open_binary` format/mapping; ESP/container: `open_binary` then `inspect_artifact`; region scan/extraction: advertised firmware tools. |
-| Archive/package inventory or extraction, PyInstaller, Bun/Deno executable, Godot PCK, UnityFS or wxapkg                                                            | [Packages and archives](references/packages-and-archives.md): `open_binary`, then `inspect_artifact`; read extraction/mount limits before materializing.                                                                                           |
-| ASAR/extracted JavaScript or Electron tree, feature trace or version comparison                                                                                    | [JavaScript applications](references/javascript-applications.md): `analyze_javascript_application`.                                                                                                                                                |
-| APK code, manifest, classes, methods or incoming references                                                                                                        | [Android applications](references/android-applications.md): `inspect_android_package`; signing/inventory use their separate routes.                                                                                                                |
-| PE/ELF/Mach-O tables or hardening, Go/Rust/IL2CPP/GPU/model metadata, Windows shortcuts, kernel config, pyc/BEAM, SQLite, WASM, EVM, APK signing or recorded crash | [Offline format inspectors](references/format-inspectors.md): select the named target-free inspector in that format's section.                                                                                                                     |
-| Shipped application exposure for a local app, install directory, ASAR or executable                                                                                | [Offline format inspectors](references/format-inspectors.md#release-exposure): `audit_release_exposure`; preserve unknown/unassessed checks and marker-only credential findings.                                                                   |
-| Apple plist, Interface Builder, asset catalog or keyed archive resources; unknown desktop installation directory                                                   | [Resource and directory inspection](references/native-and-artifacts.md#apple-app-resources): `open_bundle_resources` or `inspect_desktop_application`; no executable/provider required for resource sessions.                                      |
-| Native execution, live calls, instrumentation, process effects; browser/Electron runtime or agent-performed actions                                                | [Runtime observation](references/runtime-observation.md): select the observation and declared lifecycle first. Already-open page: `list_browser_targets`; Electron: `list_electron_targets`; agent actions: `capture_browser_scenario`.            |
-| Recorded HAR/SAZ/mitmproxy/qlog, packet capture, protocol-field discovery or logic/signal recording                                                                | [Recorded captures](references/recorded-captures.md): `inspect_web_network_capture`, packet/signal inspection or decoder discovery as appropriate; no live acquisition or recorded-URL fetch.                                                      |
-| Web AgentEnv bundle from a live instance or capture; task for an exported desktop environment                                                                      | [Environments](references/environments.md), including its desktop-task section: read before recording/export; `capture_browser_scenario` then `export_environment_bundle` for a live web bundle.                                                   |
-| Linux desktop application environment built from a user request                                                                                                    | [Desktop builds](references/desktop-build.md): drive the selected app on a build computer, then export and witness.                                                                                                                                |
-| Retained Evidence paging/composition, comparisons, unknowns or a broader investigation                                                                             | [Evidence workflows](references/evidence-workflows.md): use existing source IDs and the smallest view that answers the question.                                                                                                                   |
+```bash
+npx -y rea-agents@latest analyze-javascript-application /absolute/path/to/app --json
+```
 
-If a needed tool is unadvertised, inspect `binary_session {}` and
-`result.tool_availability` for availability reasons/remediation; never call an
-unadvertised tool. Expand group `names`, inheriting inventory client requirements
-unless overridden. Their union is the complete in-scope inventory; `tools/list`
-owns canonical schemas. The default summary keeps provider candidate status,
-code and reason; `omitted_fields` names exclusions. Request `detail: "complete"`
-only for remediation/diagnostics, candidate capabilities or tool effect flags.
-Candidate facts describe possible providers; top-level capabilities and
-`analysis_provider_binding` describe the current route. Keep those scopes distinct.
+No MCP registration or native engine is required. Read the returned Evidence,
+graph, limitations, and unknowns with the same care as an MCP result; the CLI
+returns the Evidence record directly. This fallback does not establish that
+MCP is configured. Native CLI tasks still require their selected engine.
+Prefer an existing `rea` CLI. When none is installed and REA package acquisition
+is within the authorized scope, use `npx -y rea-agents@latest` in its place from a
+trusted directory outside the target project, with absolute target paths. Managed
+setup binds the current installation directly; retain its runtime and package
+cache entry, or restore that installation before repairing its registration.
+For setup-owned registration, configuration repair starts with a read-only setup plan; show paths, backups and
+skill replacement before applying authorized changes. Preserve existing approval
+and target-execution boundaries.
+
+## Route the target first
+
+Choose the first tool from the target the user supplied. Use `open_binary` for
+active-target native or archive workflows; target-free tools take their own
+explicit path or endpoint and do not need it.
+
+- ASAR or extracted JavaScript/Electron tree:
+  `analyze_javascript_application`.
+- Archive/package member inventory (ZIP/APK/IPA/MSIX/AppX, DMG, or a PyInstaller
+  executable): `open_binary` with the supplied local path; use `inspect_artifact`
+  when its graph and findings help answer the question. A PyInstaller CArchive
+  is the application layer; the bootloader is only the native prefix. Start with
+  `detail: "summary"` for large inventories, then use retained inventory views
+  for selected member fields/pages. A complete small inventory can cost less
+  than extra calls and schemas. Runtime options are metadata, not files.
+  Compact delivery still inventories and hashes members. See the
+  [Python packaged applications guide](https://github.com/morluto/rea/blob/main/docs/python-packaged-applications.md).
+- Bun `--compile` executable: `open_binary`, then `inspect_artifact`; the Bun
+  module graph is the application layer. Extract stored modules and analyze the
+  JavaScript. See [Bun standalone executables](https://github.com/morluto/rea/blob/main/docs/bun-standalone.md).
+- Deno `deno compile` executable: `open_binary`, then `inspect_artifact`; the
+  embedded virtual filesystem is the application layer. Extract stored files and
+  analyze compatible JavaScript. Environment-file values and CA certificate bytes
+  are retained in local Evidence with their counts and byte lengths. The npm
+  snapshot length is reported and not decoded, and the executable is not run. See
+  [Deno compile executables](https://github.com/morluto/rea/blob/main/docs/deno-standalone.md).
+- Apple app plist, Interface Builder, asset catalog, or keyed archive resources:
+  `open_bundle_resources`, then the selected resource tool. This does not
+  require `CFBundleExecutable` or a deep provider. Resource sessions have no
+  executable digest or binary snapshot; use `open_binary` for native code.
+- Android APK code, classes, methods, or incoming references:
+  `inspect_android_package`, then focused Android tools when advertised.
+  Archive member inventory still uses the archive route above.
+- Managed PE/CLI assembly: `inspect_managed_artifact`.
+- Firmware image: `inspect_firmware_regions` when advertised. Use
+  `extract_firmware` when extraction is requested, with a caller-selected new
+  absolute output directory. These tools use caller-supplied Binwalk/Unblob on
+  Linux; see the [firmware guide](https://github.com/morluto/rea/blob/main/docs/guides/firmware/firmware-analysis.md).
+- ESP32-family app or bootloader image (`0xE9` magic): `inspect_artifact`
+  without a provider; see the ESP section of `references/native-and-artifacts.md`.
+- .NET NativeAOT PE/ELF: native Ghidra analysis. Native code can yield pseudocode;
+  consult the NativeAOT workflow in `references/native-and-artifacts.md` for
+  optional metadata recovery and its supported host/layout boundary.
+- Native program behavior, live calls, instrumentation, or command-run effects:
+  choose the matching [native runtime workflow](references/runtime-observation.md#native-runtime-questions).
+  `open_binary` and `trace_native_values` establish static facts; they do not
+  observe execution. Runtime tools have distinct launch, attach, mutation and
+  cleanup effects, and do not require a preliminary static session unless their
+  own contract says so.
+- User-owned browser page already open: `list_browser_targets`.
+- Passive traffic capture of a SPA: the default scope is the page's own origin,
+  and data APIs usually live on another origin. Discover API origins from
+  absolute URLs in loaded bundles or script text via JavaScript analysis tools
+  (another passive capture without those origins still scopes to the page
+  origin, and `completeness.excluded` counts lack URLs). Pass both the page
+  origin and those API origins in `allowed_origins` — the list replaces the
+  default scope — and read `completeness.excluded` before reporting that an
+  API is absent. For long captures with body shapes, use
+  `export_evidence_bundle` instead of expecting an inline result.
+- Browser action the agent must perform (click, type, navigate):
+  `capture_browser_scenario` with declared actions. See
+  [runtime observation](references/runtime-observation.md) for connect/launch
+  and the focused follow-up browser tools.
+- Agent environment (AgentEnv bundle) of a web application: read
+  [environments](references/environments.md) before recording; it records with
+  `capture_browser_scenario` and writes the bundle with `export_environment_bundle`.
+- Task for a desktop application's exported environment, from a
+  `capture-desktop-application` folder: read
+  [desktop application tasks](references/environments.md#tasks-for-a-desktop-application-environment).
+- Environment of a Linux desktop app with a task, from a request such as
+  "build me an Inkscape environment where the agent changes a poster's
+  headline": read [desktop builds](references/desktop-build.md). You drive the
+  app on a build computer from `start_build_computer`, then export and witness.
+- Fiddler Everywhere traffic or a native `.saz` capture: read the
+  [Fiddler workflow](references/runtime-observation.md#fiddler-everywhere-traffic-debugging).
+  Inspect saved SAZ with `inspect_web_network_capture` and `format: "saz"`;
+  connect Everywhere's separate MCP server for live traffic.
+- Retained HAR/SAZ/native mitmproxy/aioquic qlog capture: `inspect_web_network_capture`. Select
+  original record ordinals when useful; inspect producer fields and byte/number
+  sidecars without fetching recorded URLs or claiming live attribution. HAR/SAZ/qlog
+  old-generation heap is `REA_CAPTURE_DECODER_OLD_GENERATION_MIB` (default 192 MiB,
+  floor 8, maximum 16384), set where REA starts. Host memory and `NODE_OPTIONS` do not
+  raise it. See
+  [historical captures](https://github.com/morluto/rea/blob/main/docs/guides/web/web-network-captures.md) for the exact upstream
+  profile and credential exclusions. Qlog support starts with aioquic 1.5.0
+  JSON 0.3; stream coverage and endpoint loss declarations do not recover bodies
+  or establish network loss. Prefer detail summary, then kind qlog trace/stream
+  pages; select one trace, stream ID and direction for contribution/range pages.
+  Facts retain all-recorded-events-before-selection scope. Other qlog layouts
+  and PCAP correlation are unsupported.
+- User-owned Electron runtime already open: `list_electron_targets`. Before an electron-main Inspector attempt, check packaged fuses with `inspect_electron_fuses`; disabled `EnableNodeCliInspectArguments` means the app ignores `--inspect`.
+- Shipped application exposure for a local `.app`, install directory, ASAR, or executable: `audit_release_exposure` / `rea audit-release-exposure`. Each check has a stable id, a status, and fixed remediation. Unknown and unassessed checks stay in the report. This does not establish that an application is secure. Credential-marker hits report the rule id and location only. See [release exposure](https://github.com/morluto/rea/blob/main/docs/release-exposure.md).
+- Packet-field discovery: `inspect_packet_fields` searches installed protocols
+  and field names/types before choosing fields for `inspect_packet_capture`.
+  Query/page the relevant records; registrations do not establish observed traffic.
+- Recorded packet captures: `inspect_packet_capture` with selected TShark fields,
+  display filters and decode-as rules. Requires Linux, util-linux prlimit and
+  TShark 4.x on an absolute `PATH` entry or via the optional absolute
+  `REA_TSHARK_COMMAND` override; see
+  [capture guide](https://github.com/morluto/rea/blob/main/docs/guides/data/packet-captures.md).
+  Packet limits count input packets before filtering. Output defaults to 50 matching
+  frames; use `match_offset` / `match_limit` for later pages, or explicit null
+  limit for complete matching output. Preserve the source digest across pages.
+  This does not perform live capture.
+  For address/DNS/SNI/protocol summaries use `inspect_pcap_capture` with MCP
+  detail summary, then retained kind pcap pages. Requires Wireshark 4.x tshark
+  and capinfos, with Linux prlimit or Windows Job Objects; see
+  [packet summaries](https://github.com/morluto/rea/blob/main/docs/guides/data/pcap-analysis.md).
+  Select an exact DNS question for address pages; page exhaustion does not prove
+  network completeness. This reads recordings and does not acquire traffic.
+- Decoder discovery: `inspect_signal_decoders` searches installed decoder IDs
+  or selected decoder help, including channels and options. Query and page the
+  relevant details before choosing settings; this does not scan hardware.
+- Recorded logic/signal traces: `decode_signal_recording` with explicit decoder,
+  channel and protocol settings. On Linux with util-linux prlimit, supply sigrok-cli 0.7.2 through
+  `REA_SIGROK_COMMAND`; see [recording guide](https://github.com/morluto/rea/blob/main/docs/guides/signals/signal-recordings.md).
+  Output defaults to 50 literal lines; select annotation classes and page with
+  `line_offset` / `line_limit`, or explicit null for all remaining output.
+  Window counts are display lines, not annotations or samples. Preserve source digests across pages.
+  This reads existing files; it does not acquire hardware signals.
+- Explicit local WebAssembly artifact: `inspect_wasm_artifact`. Supply WABT 1.0.42
+  through `REA_WABT_BIN_DIRECTORY`; retain exact byte/tool identity and decoded WAT.
+  Select JavaScript glue and candidate paths explicitly and preserve URL/path
+  ambiguity. No implicit fetch or target execution; WAT is not original source.
+  See the [WASM guide](https://github.com/morluto/rea/blob/main/docs/guides/wasm/wasm-artifacts.md).
+- Explicit local Android APK (or ZIP/AAB) signing: `inspect_apk_signature`.
+  Prefer its summary or selected signer/digest/lineage page; retain complete
+  Evidence and use `inspect_analysis_view` kind `apk-signature` for follow-ups.
+  Reports which schemes are present (v1 JAR, v2, v3, v3.1), each signer's
+  certificate SHA-256 fingerprint (v1 included, from its PKCS#7 block), v3 key-rotation lineage, source stamp, and
+  whether each signed digest still matches the bytes (self-consistency). This
+  is identity and integrity, not a trust or authenticity verdict.
+  See the [APK signature guide](https://github.com/morluto/rea/blob/main/docs/apk-signature.md).
+- Explicit local EVM bytecode carrier: `inspect_evm_interface` with caller-selected
+  `raw` or `hex` encoding. The owned worker deadline is `timeout_ms` (default
+  30000 ms, range 1..2147483647, next call). Select byte, address-space and CPU
+  allowances separately through `producer_limits`; tighter inherited OS limits
+  remain enforced, and the 256 MiB worker heap is independent. Preserve
+  carrier/decoded digests and treat selectors,
+  argument strings and mutability as inferred candidates; this performs no chain
+  lookup or target execution. For the complete inferred interface without hex or
+  duplicate stdout, discover support and select `detail: "summary"`; complete
+  Evidence stays retained under its original identity. Use the returned
+  `parent_evidence_id` for exact function or `bytecode_hex`/stdout/stderr pages.
+  Candidate coverage does not establish interface completeness. See the
+  [EVM bytecode guide](https://github.com/morluto/rea/blob/main/docs/guides/evm/evm-bytecode.md).
+- Embedded Go compiler, module, build-setting metadata, or static `(ptr, len)`
+  string literals: `inspect_go_binary` with the explicit local file (pass
+  `include_string_literals: true` and optional `max_string_literals` for bounded
+  inferred static string-header recovery, and page retained rows with
+  `inspect_analysis_view` kind `go` collection `string_literals`). This built-in
+  target-free inspection requires no Go installation or disassembler. Preserve
+  raw module bytes and source ranges; absent metadata does not prove a binary is
+  not Go, static string recovery excludes compiler-inlined stack or immediate
+  strings, and unparsed text records remain explicit. See the
+  [Go binary guide](https://github.com/morluto/rea/blob/main/docs/guides/native/go-binaries.md).
+- Embedded CUDA/HIP code objects: `inspect_gpu_code_objects`; inventory targets
+  first, then decode selected payloads and query retained GPU Evidence.
+- Bundled GGUF or safetensors model declarations: `inspect_model_artifact`
+  with the explicit local file. Default header hashing leaves full-file identity
+  unknown; `hash: "full"` additionally reads all weights for SHA-256. MCP returns
+  retained summaries; optional `view` selects one page in the initial call. Use
+  `inspect_analysis_view` kind `model-artifact` for
+  batched metadata keys, tensor names or type totals, selected fields and pages.
+  Preserve header-scoped indices and array preview omissions; do not infer model
+  behavior from declared architecture or tensor types. See the
+  [model file guide](https://github.com/morluto/rea/blob/main/docs/model-artifacts.md).
+- Unity IL2CPP names, member owners and string literals from `global-metadata.dat`:
+  `inspect_il2cpp_metadata` with the explicit local file; MCP defaults to compact
+  counts. Query sections, images, types, methods, fields, properties, events,
+  parameters or string_literals through `inspect_analysis_view`
+  with the returned `parent_evidence_id`, using exact filters, batch indices and
+  selected fields. Parameters accept `method_indices`, avoiding a scan of all
+  parameter pages. Missing collections in historical Evidence remain unknown.
+  Versions 24, 27, 29, 31 and 39; no Unity, IL2CPP
+  binary or disassembler is run. Unrecognized magic (possibly encrypted) and
+  other versions are unsupported. Optional matching `binary_path` enables static native addresses and type references; query `method_addresses` with indices or addresses/states, preserving every shared-body method and both artifact identities. Without a linked binary native counts remain unknown. Byte-pattern fallback is inferred; no runtime behavior or same-build pairing is proved. See the
+  [IL2CPP metadata guide](https://github.com/morluto/rea/blob/main/docs/il2cpp-metadata.md).
+- Embedded cargo-auditable Rust dependency metadata: `inspect_rust_binary`
+  with the explicit local ELF, PE, or thin Mach-O file. This built-in
+  target-free inspection requires no Rust installation or disassembler. Preserve
+  package order, dependency indices, and the section file range. A missing
+  `.dep-v0` section is null metadata, not proof the binary has no dependencies.
+  A rustc commit hash is an inference from a read-only `/rustc/<40-hex>/` path.
+  For selected crates, use `view: {collection: "packages", names: [...],
+  fields: ["name", "version", "source"]}` on the first call; explicitly include
+  `dependencies` for direct index references. Use `detail: "summary"` for counts
+  and ranges, or retained `inspect_analysis_view` Rust package/edge pages for
+  subsequent filters. Keep the complete parent Evidence accessible; CLI compact
+  delivery requires `--evidence_output`.
+  See the
+  [Rust binary guide](https://github.com/morluto/rea/blob/main/docs/rust-binaries.md).
+- PE Authenticode certificate table: `inspect_pe_certificates` with the explicit
+  local PE32/PE32+ file. This built-in target-free inspection requires no
+  Windows SDK. Preserve WIN_CERTIFICATE ranges, types, payload SHA-256, and
+  PKCS#7 signer/X.509 identity fields, including `subject_key_identifier_hex`
+  when the signer uses that form. A missing security directory is observed
+  absence of that table, not proof the image is unsigned in every catalog sense.
+  Signatures are not verified. See the
+  [PE certificate guide](https://github.com/morluto/rea/blob/main/docs/pe-certificates.md).
+- PE import and delay-load tables: `inspect_pe_imports` with the explicit
+  local PE32/PE32+ file. This built-in target-free inspection requires no
+  Windows SDK. Preserve DLL names, named and ordinal functions, ILT/IAT RVAs,
+  delay-load descriptors, and the pefile import hash of the standard import
+  table. A missing import directory is observed absence of that table, not
+  proof the image never imported APIs through another table. Bound snapshots
+  and loaded IAT contents remain unobserved. See the
+  [PE import guide](https://github.com/morluto/rea/blob/main/docs/pe-imports.md).
+- Windows Shell Link shortcuts: `inspect_windows_shortcut` with the explicit
+  local `.lnk` file. This built-in target-free inspection requires no Windows
+  SDK or COM. Preserve header flags and FILETIMEs, IDList items, LinkInfo
+  volume and network names, StringData arguments and icon location, ExtraData
+  signatures, and Tracker droid facts labelled derived. Unknown IDList items
+  and ExtraData signatures stay raw with SHA-256. The target path is not
+  resolved or executed. See the
+  [Windows shortcut guide](https://github.com/morluto/rea/blob/main/docs/guides/native/windows-shortcut.md).
+- PE TLS callbacks and load configuration: `inspect_pe_load_config` with the
+  explicit local PE32/PE32+ file. This built-in target-free inspection requires
+  no Windows SDK. Preserve callback VAs as strings with their RVA, file offset
+  and section, and the size-covered security cookie, SafeSEH, CFG,
+  EH-continuation and XFG fields. Guard-flag mitigation claims are inferred
+  linker declarations, not proof of enforcement. A missing directory is observed
+  absence of that declaration. See the
+  [PE load configuration guide](https://github.com/morluto/rea/blob/main/docs/pe-load-config.md).
+- ELF, PE, and Mach-O hardening: `inspect_binary_hardening` with the explicit
+  local file. This built-in target-free inspection requires no pwntools or other
+  host tool. One image per Mach-O slice. Each property is `enabled`, `disabled`,
+  `partial`, `unknown`, or `not_applicable`, with the header field or import
+  behind it. A missing stack-protector or fortify import is `unknown`, not
+  proof the protection was omitted. Authenticode and Mach-O code signatures
+  report presence, not validity. Hardened runtime stays unknown and names
+  `inspect_signature`. The image is not executed. See the
+  [binary hardening guide](https://github.com/morluto/rea/blob/main/docs/binary-hardening.md).
+- ELF note tables: `inspect_elf_notes` with the explicit local ELF32/ELF64
+  file. This built-in target-free inspection requires no disassembler. Preserve
+  PT_NOTE and SHT_NOTE ranges, owners, types, descriptor SHA-256, and GNU
+  build-id or ABI-tag fields. A missing table is observed absence of those
+  declarations, not proof the image never carried notes. See the
+  [ELF notes guide](https://github.com/morluto/rea/blob/main/docs/elf-notes.md).
+- Code that runs before `main` or at exit in ELF or Mach-O:
+  `inspect_native_initializers` with the explicit local file (and
+  `architecture` for a universal Mach-O). This built-in inspection requires no
+  disassembler. Preserve slot and target addresses, the `resolution` of each
+  entry, and its limitations. An `unresolved` slot is not a zero pointer, and a
+  missing table is not proof that no constructor exists. See the
+  [initializer guide](https://github.com/morluto/rea/blob/main/docs/guides/native/native-initializers.md).
+
+- ELF unwind function ranges: `inspect_elf_unwind_ranges` with the explicit
+  local ELF32/ELF64 file. This built-in target-free inspection requires no
+  disassembler. Preserve FDE start, end, size, containing section, exact
+  symbol name, CIE index, LSDA, and personality. Page with `offset` and
+  `limit`. A missing `.eh_frame` is observed absence of that declaration,
+  not proof the image has no functions. `.debug_frame` is reported present
+  or absent and is not decoded. CFA instructions are counted and not
+  evaluated. See the
+  [ELF unwind ranges guide](https://github.com/morluto/rea/blob/main/docs/guides/native/elf-unwind-ranges.md).
+- Apple `.ips` crash reports: `inspect_apple_ips` with the explicit local
+  file. This built-in target-free inspection requires no Xcode, atos, or
+  pwntools. Preserve first-line metadata, `bug_type` as a string, process and
+  exception facts, lossless integer registers, usedImages UUIDs, and
+  documented classification matches. Only `bug_type` `"309"` yields a crash
+  body; other types, including `"288"` stackshots, return no crash result.
+  Select an optional `join_path` for UUID-and-architecture matching to one
+  local Mach-O or dSYM; preserve slice identity and unjoined reasons. Joined
+  static addresses do not resolve function names or dSYM symbols. See the
+  [Apple IPS guide](https://github.com/morluto/rea/blob/main/docs/apple-ips.md).
+- Linux kernel module metadata: `inspect_kernel_module` with the explicit local
+  ELF32/ELF64 relocatable file. This built-in target-free inspection requires
+  no Linux host or `modinfo`. Preserve `.modinfo` key order and duplicate keys,
+  `__versions` CRC hex, section file ranges, and the appended signature
+  trailer's PKCS#7 SHA-256. A missing section is observed absence when section
+  names are known, not proof the module never carried that metadata.
+  `struct module` is not decoded. The signature is not validated. gzip, xz, and
+  zstd containers are unsupported; decompress into a separate file with an
+  external codec utility, then inspect the decompressed `ET_REL` image. See the
+  [kernel module guide](https://github.com/morluto/rea/blob/main/docs/kernel-modules.md).
+- Linux kernel version and build config (`CONFIG_*` options such as
+  `CONFIG_SECURITY_SELINUX`): `inspect_kernel_build_config` with the explicit
+  local `vmlinux`, raw `Image`, x86 `bzImage`, or `configs.ko`. This built-in
+  target-free inspection requires no Linux host or `extract-ikconfig`. Preserve
+  the banner version, config SHA-256, option count, and value kinds; use `name`
+  for one option. An `IKCFG_ST` string is a config only with a valid gzip member
+  and `IKCFG_ED`. An absent config may mean the kernel was built without
+  `CONFIG_IKCONFIG`. Non-gzip payloads are unsupported; decompress externally,
+  then inspect the result. See the
+  [kernel build config guide](https://github.com/morluto/rea/blob/main/docs/guides/native/kernel-build-config.md).
+- CPython 3.7–3.13 `.pyc` header, nested marshal code objects and bytecode
+  identity: `inspect_python_pyc` with the explicit local file. This built-in
+  target-free inspection requires no Python installation. Preserve magic,
+  PEP 552 hash or mtime/size, nested names, and bytecode SHA-256. Bytecode is
+  not decompiled, and a decoded header does not prove the matching `.py` is
+  present. See the
+  [CPython pyc guide](https://github.com/morluto/rea/blob/main/docs/python-pyc.md).
+- Erlang BEAM IFF container: `inspect_erlang_beam` with the explicit local
+  `.beam` file. This built-in target-free inspection requires no Erlang
+  installation. Preserve chunk names, atom order, export/import/lambda tables,
+  and the Code header. For selected facts, prefer `detail: "summary"` plus a
+  `beam` view (collection, names/indices, fields and paging), retaining the full
+  Evidence. CLI summary requires `--evidence_output`. Missing table counts stay
+  null; decoded empty tables count zero. Bytecode is not executed. See the
+  [Erlang BEAM guide](https://github.com/morluto/rea/blob/main/docs/erlang-beam.md).
+
+- Explicit local SQLite database snapshot: `inspect_sqlite_database`. Inspect
+  with Node.js 24.x >=24.15 or 26+ providing native SQLite limits. Read schema
+  first (compact inventory by default), or select an exact ordinary table
+  and `columns` when rows are needed. Defaults are selected-table schema and
+  20 rows; request `schema_scope: "database"` for the complete stored schema. Preserve both database and existing sibling WAL identity; row samples
+  report truncation and have unspecified order. Use a quiescent snapshot rather
+  than assuming a live database/WAL pair was copied atomically. See the
+  [SQLite guide](https://github.com/morluto/rea/blob/main/docs/guides/data/sqlite-databases.md).
+- Explicit Linux ELF file for offline layout, symbols, relocations or static
+  mitigation evidence: `inspect_binary_layout`. This target-free operation uses
+  caller-supplied pwntools without opening a disassembler database. Preserve its
+  raw locations and inference/coverage limits; see the
+  [offline binary guide](https://github.com/morluto/rea/blob/main/docs/guides/native/binary-diagnostics.md).
+- Supplied Linux x86-64 ELF core: `inspect_recorded_crash`. Read every recorded
+  thread, raw note and source-bound register without opening a live target.
+  Historical PIDs do not select live processes. Request `include_debugger_context`
+  only when core-only mapping candidates help; see the
+  [recorded crash guide](https://github.com/morluto/rea/blob/main/docs/guides/native/recorded-crashes.md).
+- PE resource type/name/language trees and static icon group references:
+  `inspect_pe_resources` with an explicit path. This portable operation requires
+  no active target or engine; opaque payloads retain file ranges and SHA-256.
+- PE export directories and ELF dynamic-symbol exports: `inspect_native_exports`
+  with an explicit path; portable, no active target or engine.
+- Native executable, library, or analysis database: `open_binary`, then
+  use focused analysis tools directly; call `binary_overview` when metadata or
+  inventory context is useful and available from the selected provider.
+
+For an existing IDA MCP configuration, use `open_binary` with the original
+input binary and `provider_id: "ida"`; do not pass an `.idb` or `.i64` database.
+The legacy attached profile binds the already-open GUI input by SHA-256 and
+leaves its database open. The modern headless profile creates and closes an
+owned private database without saving. Installation and registration reuse
+[mrexodia/ida-pro-mcp](https://github.com/mrexodia/ida-pro-mcp); see the
+[IDA provider guide](https://github.com/morluto/rea/blob/main/docs/providers/ida-provider.md).
+For migration to the separate native IDA MCP plugin, follow the guide's
+[legacy attached migration](https://github.com/morluto/rea/blob/main/docs/providers/ida-provider.md#migrate-a-legacy-attached-registration).
+Replace the upstream registration explicitly; modern mrexodia tools are not the
+native profile. Native attachment preserves the database but does not provide
+direct callers; check session availability and the guide's verification limits.
+Proceed directly to `analyze_function`, pseudocode, or function/string searches
+after opening; IDA does not supply `binary_overview`. Consult session availability
+when composing broader workflows. Modern direct callers and unsupported dossier
+facets remain unknown; live IDA results are not replayed from snapshots.
+
+With `REA_KUNA_COMMAND` configured, `provider_id: "kuna"` answers procedure
+lists, lookups, pseudocode, assembly and `analyze_function` from Kuna's cached
+whole-program export. Callers/callees use a separate body-free graph shared by
+the session; `xrefs` returns exact-address incoming sites and filters forwarding
+aliases. These queries and inventory lookups do not trigger a C export. Indirect
+targets remain unknown and can make relationship coverage partial. Reference
+caches expire on close; a fresh CLI process repeats the native reference walk.
+Typed references, CFG, raw-memory tools and annotations remain unsupported. A cold first
+answer can come from a single-function run (its limitations say so); assembly
+waits for the export. For pseudocode and assembly, `raw_result.source` names the
+cached `.c`/`.asm` file and byte span, which you may search directly instead of
+calling per function. See the
+[Kuna provider guide](https://github.com/morluto/rea/blob/main/docs/kuna-provider.md).
+
+If the user says only "my app" without a name, path, or target already selected
+for this task, ask which app before searching for artifacts. Paths in unrelated
+saved reports and nearby test fixtures do not select the user's target. Resolve
+an actual human-readable app name to one clear installed artifact when possible;
+ask when matches are ambiguous. Never choose an example app on the user's behalf.
+
+In a target-free session, use `open_binary` to bind any archive/package or
+native target whose analysis tool operates on the active target. Do not call an
+unadvertised tool; inspect `binary_session` with `{}` and its
+`result.tool_availability` for availability reasons and remediation. Expand each
+group's `names` into entries, inheriting the inventory-level client requirements
+unless the group overrides them. The grouped name union is the complete tool
+inventory; use `tools/list` for canonical tool schemas. The default summary keeps
+each provider candidate's availability and target-support status, code and
+reason; `omitted_fields` lists what it leaves out. Request `{"detail": "complete"}`
+only when you need an unavailable candidate's remediation, provider
+diagnostics, candidate capability sets, or tool surface, effect and annotation
+flags; in those closed flag vocabularies, listed
+flags are true and omitted known flags are false. Provider candidate facts
+describe possible providers, while top-level session capabilities describe the
+current route and `analysis_provider_binding` names the selected provider. Do
+not treat those scopes as interchangeable.
+
+For editable Ghidra analysis that must survive a later agent process, select
+the [owned Program checkpoint workflow](references/native-and-artifacts.md#continue-editable-ghidra-analysis-across-processes)
+before making metadata edits. It is separate from Evidence/Unknown persistence.
+
+For a Ghidra startup timeout on a large native target, follow
+[deadline recovery](references/native-and-artifacts.md#ghidra-startup-deadline-recovery)
+before abandoning deep analysis or repeating the same cold import. The default
+330 seconds can be changed for new startup attempts through
+`update_analysis_settings` using `ghidra.startup_timeout_ms`. Changing the
+server's launch environment instead requires a new REA process.
 
 ## Work summary-first
 
-Start with inline Evidence and graph context. For unknown/large JavaScript or ELF
-layout, select advertised `detail: "summary"`; MCP `analyze_function` already
-defaults to a summary. Small complete results can cost less than extra pages.
-Make a focused follow-up only for a specific unanswered question. Reuse matching
-session state and Evidence instead of repeating analysis; recheck state when a
-lifecycle change makes it useful.
+Start with the default result and use its inline Evidence and graph context.
+For a JavaScript application or ELF layout whose size is unknown or large,
+call `analyze_javascript_application` or `inspect_binary_layout` with
+`"detail": "summary"`, then inspect the retained `parent_evidence_id` with
+`inspect_analysis_view`. Over MCP, `analyze_function` is summary-first without
+a flag: its default reply is the summary, and the complete dossier stays
+retained under `parent_evidence_id` for facet paging. Pass that ID as a
+`retained-evidence` reference to `compare_functions`, or inside
+`match_native_functions` `left_observations`/`right_observations` or
+`build_call_path.functions`; request `"detail": "complete"` only when the
+dossier must travel inline.
+Use the same exact-reference form in other advertised logical Evidence inputs,
+including optional sources, arrays and nested sources across domains. Each
+source can independently be inline or retained; preserve its source identity,
+coverage, partial facts and unknowns. A summary or inspection page has its own
+Evidence ID: use `normalized_result.parent_evidence_id` when the consumer needs
+the complete source. For portable CLI follow-ups, commands advertising
+`--evidence-bundle` resolve references from that selected exported bundle.
+See [Evidence composition](references/evidence-workflows.md) for scope and
+recovery; file selectors are supported only where explicitly advertised.
+Page module identities and select the returned `node_id` before requesting an
+item; a single file path can represent several graph roles. Do not repeat an
+identical analysis or rejected call without a changed input or resolved blocker.
+Recheck session state when a lifecycle change makes that status useful.
+Make a focused follow-up only when the
+returned result leaves a specific question unanswered.
 
-Complete source Evidence stays accessible through advertised retained references.
-A summary/page has its own Evidence ID; use `normalized_result.parent_evidence_id`
-when a consumer needs the complete source. Pass `{kind: "retained-evidence",
-evidence_id}` in inputs advertising that form; do not resend complete records.
-Sources may independently be inline or retained. Read the Evidence guide for
-exact filters, bundle-based CLI reuse, comparison roles and missing-source recovery.
+If MCP reports `resource_constraint` with `details.resource: "transport"`, call
+`inspect_analysis_view` with its reported same-session Evidence reference for a
+summary, one section/module, or a stable page. Use `trace_application_feature`
+once a seed is known, or call `export_evidence_bundle` to write the complete
+session to a selected path. The analysis remains complete in the ledger; avoid
+repeating it merely to request the same oversized response. Complete CLI JSON
+output also streams without a single MCP frame.
 
-For REA transport refusal or client-truncated output, inspect a smaller retained
-view or export complete Evidence. Do not parse a preview as canonical JSON or
-repeat the producer. Never invent an Evidence ID; if unknown, find it with
-`get_evidence_bundle` summary filters. Retention is bounded and session-local;
-it does not establish automatic persistence or complete source coverage.
+When the retained parent ID is unknown, use `get_evidence_bundle` with
+`detail: "summary"` and exact filters for operation, target SHA-256,
+analysis-profile digest or canonical procedure address. Select its returned
+Evidence ID for the next view instead of requesting the complete bundle. When
+that record has no supported view (for example an annotation), read it with
+`detail: "records"` and an `evidence_id` filter; `missing_evidence_links`
+names linked sources this session does not retain.
+`retention: "complete-record"` means the whole record is stored; native
+value-flow truncation and unknown metadata remain separate. For an exported
+portable bundle, use `rea inspect-evidence-bundle` with JSON containing `path`,
+`detail: "summary"` and the same `filters`. Reading that file still requires
+validating its full contents within the local memory budget.
 
-For broad investigations, read the Evidence guide's staged workflow. Track the
-questions, source boundaries and unresolved gaps; do not claim completion while
-required questions remain open. Inspect packaged resources alongside code when
-relevant, and use runtime evidence only within the requested scope.
+A client can also truncate a successful tool result without a REA transport
+error. Treat that preview as incomplete; use the retained parent ID for smaller
+views or export when the full record is needed. Do not parse the preview as
+canonical JSON or repeat the producer. For JavaScript, read the
+[module-view and trace recipes](references/javascript-applications.md) before
+drilling into a large graph; a broad bidirectional trace can expand unrelated
+containment relationships even after summary analysis.
+
+Every conclusion must distinguish observations, inferences, and unknowns. Cite
+Evidence IDs, preserve limitations and incomplete coverage, and never imply
+that static analysis observed execution. Runtime requests execute the declared
+target and lifecycle; do not broaden the target or action beyond those fields.
+
+Within the user's requested investigation, call available analysis tools
+directly. REA does not require permission grants or per-call approval flags.
+The agent client can independently require approval. Follow the declared request
+scope and the host's actual access requirements; use the recovery guide for an
+observed client-policy rejection.
+
+## Plan broader investigations
+
+For requests that span multiple features or subsystems, use a staged workflow:
+
+1. Turn the request into a checklist of questions and the evidence each answer
+   needs. Resolve target identity and constraints from the conversation and
+   workspace before asking for information again.
+2. Inspect the current REA session, artifact identity, saved analysis database,
+   bookmarks, and prior evidence. Reuse matching state; do not open duplicate
+   sessions or repeat identical analysis.
+3. Start with the smallest useful overview or inventory. Follow each question
+   from its entry point through relevant data and state changes to its result.
+   Batch related operations around a specific hypothesis, then expand only when
+   the returned evidence leaves a concrete gap.
+4. Inspect relevant packaged resources and configuration alongside code when
+   they affect the question. Use format-aware inventory and parsers; do not
+   infer behavior from filenames, strings, or layout alone.
+5. Corroborate a conclusion with the evidence type it requires. Use runtime
+   observation only when static evidence cannot answer the question and the
+   required host runtime and OS access are available. For JavaScript behavior,
+   run probes against the actual app through the available browser, Electron,
+   or process capture workflow.
+6. Decompose work into independent questions. When parallel workers are
+   available and the questions do not depend on one another, assign distinct
+   scopes, point workers to existing evidence, and ask them to return sources,
+   conclusions, and unresolved gaps. Otherwise, work sequentially.
+7. Keep a concise finding ledger linking each conclusion to Evidence IDs,
+   confidence/evidence type, search boundary, and remaining unknowns. Update
+   the shared index or investigation report so later passes can reuse results.
+
+Before finishing, revisit the original checklist. Mark each question as
+answered, partially answered, or unresolved based on its evidence; keep
+bounded negative searches bounded, and do not describe a broad investigation
+as complete while required questions remain open.
+
+## Read only the relevant guide
+
+- Native binaries, managed assemblies, archives, and extraction:
+  [references/native-and-artifacts.md](references/native-and-artifacts.md)
+- ASARs, extracted JavaScript, feature tracing, and version comparison:
+  [references/javascript-applications.md](references/javascript-applications.md)
+- Android APK declarations, classes, methods, and static references:
+  [references/android-applications.md](references/android-applications.md)
+- Native instrumentation, debugger and process capture; browser/Electron
+  observation; static/runtime comparison:
+  [references/runtime-observation.md](references/runtime-observation.md)
+- Evidence paging, comparisons, residual unknowns, and verification:
+  [references/evidence-workflows.md](references/evidence-workflows.md)
+- Agent environments (AgentEnv bundles) of a web application, recorded from
+  a live instance or a HAR:
+  [references/environments.md](references/environments.md)
+- Tasks for a desktop application's exported environment:
+  [references/environments.md](references/environments.md#tasks-for-a-desktop-application-environment)
+- A desktop app environment from a request, built and witnessed on build
+  computers: [references/desktop-build.md](references/desktop-build.md)
 
 ## Finish the task
 
-Explain findings in plain language and tie them to returned Evidence. For builds,
-use normal coding tools and separate observed behavior from design choices.
-Close an opened native/resource session with `close_binary` when no longer needed.
-Evidence references and Unknown history remain on the same MCP connection.
-End with `reset_investigation`; use `save_path` to checkpoint the complete bundle
-before clearing. Import that bundle, including its source closure, to restore
-original references and Unknown history in a new process. Editable Ghidra Program
-checkpoints are separate; read that workflow before edits needing persistence.
+Explain findings in plain language and tie them to returned evidence. When the
+user asks to build something, use normal coding tools and separate observed
+behavior from design choices. Close an opened native session with
+`close_binary` when its engine is no longer needed. Evidence references and Unknown
+history remain in memory on the same MCP connection. End the investigation with
+`reset_investigation`; supply `save_path` to checkpoint the complete bundle before
+clearing. Import the saved bundle to restore original references and Unknown
+history in a new process. Retention is bounded and is not automatic persistence.

.agents/skills/reverse-engineer-anything/references/connection-and-recovery.md

@@ -1,23 +1,11 @@
 # Connection and task recovery
 
-Read only the sections needed for the selected task.
-
-- [Choose the remedy from the observed failure](#choose-the-remedy-from-the-observed-failure)
-- [Distinguish client and server deadlines](#distinguish-client-and-server-deadlines)
-- [Recovery ladder](#recovery-ladder)
-- [REA installed as a client plugin](#rea-installed-as-a-client-plugin)
-- [Repair registration when needed](#repair-registration-when-needed)
-- [Codex tool approval is separate from REA readiness](#codex-tool-approval-is-separate-from-rea-readiness)
-- [Repair an editable REA limit](#repair-an-editable-rea-limit)
-- [Continue through an installed CLI](#continue-through-an-installed-cli)
-
 Read this guide only for an observed connection, command, approval or provider
 blocker. Keep the user's target and question in view. Explain what prevented the
 operation and the next useful step; after repair, continue the investigation.
 
 Commands below use an installed `rea`. If none is available, use the package
-invocation under [Continue through an installed CLI](#continue-through-an-installed-cli)
-when acquiring REA is authorized. Follow the connected
+invocation in `SKILL.md` when acquiring REA is authorized. Follow the connected
 server's schemas and the installed CLI's help rather than assuming repository
 main and an older release expose the same arguments. Tool-name prefixes belong
 to the client: find the canonical REA operation in its actual tool listing and
@@ -234,25 +222,6 @@ not edit host deadlines, response buffers, registrations or fixed limits.
 
 ## Continue through an installed CLI
 
-A concrete CLI fallback for an operator-supplied JavaScript tree or ASAR is:
-
-```bash
-npx -y rea-agents@latest analyze-javascript-application /absolute/path/to/app --json
-```
-
-No MCP registration or native engine is required. Read the returned Evidence,
-graph, limitations, and unknowns with the same care as an MCP result; the CLI
-returns the Evidence record directly. This fallback does not establish that
-MCP is configured. Native CLI tasks still require their selected engine.
-Prefer an existing `rea` CLI. When none is installed and REA package acquisition
-is within the authorized scope, use `npx -y rea-agents@latest` in its place from a
-trusted directory outside the target project, with absolute target paths. Managed
-setup binds the current installation directly; retain its runtime and package
-cache entry, or restore that installation before repairing its registration.
-For setup-owned registration, configuration repair starts with a read-only setup plan; show paths, backups and
-skill replacement before applying authorized changes. Preserve existing approval
-and target-execution boundaries.
-
 Check the command environment only if the launcher fails:
 
 ```bash

.agents/skills/reverse-engineer-anything/references/environments.md

@@ -1,13 +1,5 @@
 # Agent environments from a live application
 
-Read only the sections needed for the selected task.
-
-- [Before recording](#before-recording)
-- [Record a session an environment can learn writes from](#record-a-session-an-environment-can-learn-writes-from)
-- [Check coverage](#check-coverage)
-- [Export and hand off](#export-and-hand-off)
-- [Tasks for a desktop application environment](#tasks-for-a-desktop-application-environment)
-
 Use this when the user asks for an agent environment (an AgentEnv bundle) of a
 web application: a folder `agent-env` deploys, whose MCP tools answer from what
 one recording of the application observed and serve the writes that recording

.agents/skills/reverse-engineer-anything/references/evidence-workflows.md

@@ -1,25 +1,7 @@
 # Evidence, comparison, and verification workflows
 
-Read only the sections needed for the selected task.
-
-- [Read and compose source Evidence](#read-and-compose-source-evidence)
-- [Inventory and decompilation views](#inventory-and-decompilation-views)
-- [Unknowns and comparisons](#unknowns-and-comparisons)
-- [Process captures and environment views](#process-captures-and-environment-views)
-- [Native export table pages](#native-export-table-pages)
-- [Recover oversized delivery and find retained Evidence](#recover-oversized-delivery-and-find-retained-evidence)
-- [Plan broader investigations](#plan-broader-investigations)
-
-## Read and compose source Evidence
-
-Evidence-producing MCP tools return Evidence directly in `structuredContent`.
-CLI complete JSON returns Evidence directly. CLI summary delivery through
-`--evidence_output` returns `{summary, evidence_bundle}`: `summary` is the view's
-Evidence, and `evidence_bundle` is the export receipt with its path and byte count.
-Read `summary.normalized_result.parent_evidence_id` for the complete source ID;
-pass the receipt's path to `--evidence-bundle` on a later supported CLI command.
-Advertised summary delivery retains the complete source separately;
-complete inline delivery is available where the connected contract says so. Read `normalized_result` for operation data,
+Evidence-producing tools return the complete Evidence record directly in MCP
+`structuredContent` and CLI JSON. Read `normalized_result` for operation data,
 and cite `evidence_id` when another tool or your explanation needs a stable
 reference. Preserve authority, limitations, coverage, and residual unknowns. If a result
 is incomplete or paginated, continue only when the remaining data matters to
@@ -67,8 +49,6 @@ An input bundle grants no implicit file lookup: `evidence-file` and historical
 graph file selectors are accepted only by contracts advertising them. See the
 [CLI reuse guide](https://github.com/morluto/rea/blob/main/docs/getting-started/cli.md#import-export-and-compare-evidence).
 
-## Inventory and decompilation views
-
 Completed `inspect_artifact` inventory also supports a summary and
 `view: {kind: "inventory", collection: "occurrences", logical_paths: ["package.json", "main.js"]}`.
 Omit `logical_paths` to page all available occurrences; `[]` selects none.
@@ -101,8 +81,6 @@ exact `where.indices`, `addresses`, `procedure_addresses` or `statuses`. Add
 `collection: "pseudocode"` with one row's `index` to page a large body. Batch
 limitations aggregate every item, so do not attribute them to one successful row.
 
-## Unknowns and comparisons
-
 Use `record_unknown` to track unresolved questions in the user's investigation
 and name the authority or environment still required. Supply supporting and
 contradicting evidence IDs. It updates the session registry directly.
@@ -133,8 +111,6 @@ or absence:
 - `verify_reconstruction` evaluates a finite typed specification. A pass covers
   only declared comparable claims, not global equivalence.
 
-## Process captures and environment views
-
 Process captures are opt-in behavioral evidence, not a security sandbox. V3
 captures cannot be upgraded to V4; rerun the original scenario. Distinguish root
 exit from descendant settlement and require freshness when the task needs it.
@@ -188,8 +164,6 @@ catalog and state size checks. `rea environment-build <capture.har> <dir>`
 inspects, projects and exports in one step and prints the same coverage, or
 `null` when the bundle serves replay.
 
-## Native export table pages
-
 For a PE/ELF export question, `inspect_native_exports` accepts `detail: "summary"`
 and retains the complete table. Query `inspect_analysis_view` with
 `view: {kind: "native-exports", collection: "exports", names: ["selected_name"],
@@ -200,69 +174,3 @@ coverage describes selected retained rows; `summary.coverage` or `exports.covera
 still describes the source table, so an empty partial page does not prove absence.
 CLI `inspect-native-exports --detail summary --evidence_output PATH` saves the
 complete bundle for later `inspect-analysis-view --evidence-bundle PATH` requests.
-
-## Recover oversized delivery and find retained Evidence
-
-If MCP reports `resource_constraint` with `details.resource: "transport"`, call
-`inspect_analysis_view` with its reported same-session Evidence reference for a
-summary, one section/module, or a stable page. Use `trace_application_feature`
-once a seed is known, or call `export_evidence_bundle` to write the complete
-session to a selected path. The analysis remains complete in the ledger; avoid
-repeating it merely to request the same oversized response. Complete CLI JSON
-output also streams without a single MCP frame.
-
-When the retained parent ID is unknown, use `get_evidence_bundle` with
-`detail: "summary"` and exact filters for operation, target SHA-256,
-analysis-profile digest or canonical procedure address. Select its returned
-Evidence ID for the next view instead of requesting the complete bundle. When
-that record has no supported view (for example an annotation), read it with
-`detail: "records"` and an `evidence_id` filter; `missing_evidence_links`
-names linked sources this session does not retain.
-`retention: "complete-record"` means the whole record is stored; native
-value-flow truncation and unknown metadata remain separate. For an exported
-portable bundle, use `rea inspect-evidence-bundle` with JSON containing `path`,
-`detail: "summary"` and the same `filters`. Reading that file still requires
-validating its full contents within the local memory budget.
-
-A client can also truncate a successful tool result without a REA transport
-error. Treat that preview as incomplete; use the retained parent ID for smaller
-views or export when the full record is needed. Do not parse the preview as
-canonical JSON or repeat the producer. For JavaScript, read the
-[module-view and trace recipes](javascript-applications.md) before
-drilling into a large graph; a broad bidirectional trace can expand unrelated
-containment relationships even after summary analysis.
-
-## Plan broader investigations
-
-For requests that span multiple features or subsystems, use a staged workflow:
-
-1. Turn the request into a checklist of questions and the evidence each answer
-   needs. Resolve target identity and constraints from the conversation and
-   workspace before asking for information again.
-2. Inspect the current REA session, artifact identity, saved analysis database,
-   bookmarks, and prior evidence. Reuse matching state; do not open duplicate
-   sessions or repeat identical analysis.
-3. Start with the smallest useful overview or inventory. Follow each question
-   from its entry point through relevant data and state changes to its result.
-   Batch related operations around a specific hypothesis, then expand only when
-   the returned evidence leaves a concrete gap.
-4. Inspect relevant packaged resources and configuration alongside code when
-   they affect the question. Use format-aware inventory and parsers; do not
-   infer behavior from filenames, strings, or layout alone.
-5. Corroborate a conclusion with the evidence type it requires. Use runtime
-   observation only when static evidence cannot answer the question and the
-   required host runtime and OS access are available. For JavaScript behavior,
-   run probes against the actual app through the available browser, Electron,
-   or process capture workflow.
-6. Decompose work into independent questions. When parallel workers are
-   available and the questions do not depend on one another, assign distinct
-   scopes, point workers to existing evidence, and ask them to return sources,
-   conclusions, and unresolved gaps. Otherwise, work sequentially.
-7. Keep a concise finding ledger linking each conclusion to Evidence IDs,
-   confidence/evidence type, search boundary, and remaining unknowns. Update
-   the shared index or investigation report so later passes can reuse results.
-
-Before finishing, revisit the original checklist. Mark each question as
-answered, partially answered, or unresolved based on its evidence; keep
-bounded negative searches bounded, and do not describe a broad investigation
-as complete while required questions remain open.

.agents/skills/reverse-engineer-anything/references/firmware-and-raw-images.md

@@ -1,76 +0,0 @@
-# Firmware and raw images
-
-## Firmware regions and extraction
-
-Use `inspect_firmware_regions` when advertised. Use `extract_firmware` when
-extraction is requested, with a caller-selected new absolute output directory.
-These tools use caller-supplied Binwalk/Unblob on Linux; see the
-[firmware guide](https://github.com/morluto/rea/blob/main/docs/guides/firmware/firmware-analysis.md).
-
-## Headerless raw images and firmware containers
-
-For other headerless images (embedded firmware without ELF/Mach-O/PE/MZ
-headers), open with `format: "raw-binary"`, `architecture` (`x86`, `x86_64`,
-`arm`, `arm64`, `6502`, or `m68k`), and `entry_point` (`0x` hex). File offset 0 is mapped at
-that address. Hopper and IDA refuse this interpretation; Ghidra uses
-`BinaryLoader` with a generic language: big-endian for `m68k`, little-endian
-otherwise. Big-endian ARM, Thumb-only, and Cortex-M variants are unsupported.
-`6502` selects NMOS 6502; the entry and complete file mapping must fit within
-`0x0000..0xffff`. 65C02, 65816, cartridge containers, bank switching, and mapper
-semantics are unsupported. `m68k` requires an even entry and decodes the 68040
-superset of the 68000 family; Amiga/Atari containers and vector-table entry
-discovery are unsupported. Windows Ghidra P0 admits raw images only for `6502`. See [raw binary guide](https://github.com/morluto/rea/blob/main/docs/providers/raw-binary.md).
-
-Intel HEX, S-record, UF2 and DfuSe files are containers, not raw images:
-`inspect_artifact` lists address regions (UF2 per familyID, DfuSe per element
-with its alternate setting), and `extract_artifact` writes each region. Open an
-extracted region as `raw-binary` with its start address as `entry_point`. UF2
-numbering gaps leave coverage unknown; a DfuSe suffix CRC mismatch follows
-`integrity_policy`. See [UF2 and DfuSe](https://github.com/morluto/rea/blob/main/docs/uf2-dfuse.md).
-
-## ESP images
-
-For an ESP32-family app or bootloader image, `open_binary` then
-`inspect_artifact` with `detail: "summary"`. `root_format_metadata` holds the
-observed `header` and `descriptor` (its `kind` is the app or bootloader role)
-and the derived `chip`, `checksum`, `appended_sha256` and
-`undescribed_trailing` ranges. Segment rows (`segment-NN.bin`) carry
-`load_addr`, `classification` (`mapped`, `loaded`, `not-loaded` or `unknown`,
-per ESP-IDF `should_map`/`should_load`; RAM ranges are not checked) and
-`source_location`; select them with `inspect_analysis_view` and
-`logical_paths`. Signature blocks are `unverified`; `public_key_sha256`
-matches eFuse key digests.
-
-Unknown `source_coverage` means truncation or undescribed trailing bytes, such
-as the flash segments of a `--ram-only-header` image; do not infer absence
-from it. Erased flash is described. Integrity mismatches are facts, not
-refusals. ESP8266 images are refused, and flash-encrypted images are not
-recognized. Segments are not disassembled, and RISC-V or Xtensa segments
-cannot be opened as raw-binary. See
-[ESP images](https://github.com/morluto/rea/blob/main/docs/esp-image.md).
-
-## ESP flash images
-
-A whole-flash dump opens as `esp-flash-image`. `root_format_metadata` holds
-`partition_table` (its `method`: `default`, `caller` or a derived
-`scan`), `bootloader` and `apps` ESP image facts, the `otadata` records,
-`partitions_outside_dump` and `boot_selection`. Every table has a
-`boot_selection`: its `basis` names the bootloader's branch (`no-otadata`
-without otadata, `none` when the bootloader rejects the table, `unknown`
-when a record is missing), and it lists its `assumptions`;
-`running_partition` is always `unknown`. Each image names its entry
-`path`. With `detail: "summary"` the root kind is `esp-flash-image-summary`
-and each image is a compact `summary` (chip, entry point, segment count,
-classified load ranges, signature block count, descriptor identity,
-integrity); complete detail keeps every segment. Partition rows
-(`partitions/<label>.bin`) carry the declared `type`, `subtype`, names,
-`flash_address`, `size` and flags; `encrypted` is a storage flag, not
-proof of ciphertext. `written_offset`/`written_end` bound the bytes that
-are not erased: an app slot whose facts say `erased` can still hold an older
-image, and a limitation says so. For a dump read from another flash address, call
-`open_binary` with `format: "esp-flash-image"` and `flash_offset`; an
-`esp-image` whose `flash_dump` is set names that offset in a next probe; a
-relocated table can be named with `partition_table_offset`. No table found
-means unknown, often flash encryption. Extraction writes `nvs_keys` as
-ordinary data. See
-[ESP flash images](https://github.com/morluto/rea/blob/main/docs/esp-flash-image.md).

.agents/skills/reverse-engineer-anything/references/format-inspectors.md

@@ -1,449 +0,0 @@
-# Offline format and metadata inspection
-
-Read only the sections needed for the selected task.
-
-- [Release exposure](#release-exposure)
-- [Go build metadata](#go-build-metadata)
-- [Model files](#model-files)
-- [IL2CPP metadata](#il2cpp-metadata)
-- [Rust dependency metadata](#rust-dependency-metadata)
-- [PE Authenticode certificate tables](#pe-authenticode-certificate-tables)
-- [PE import and delay-load tables](#pe-import-and-delay-load-tables)
-- [Windows Shell Link shortcuts](#windows-shell-link-shortcuts)
-- [PE TLS callbacks and load configuration](#pe-tls-callbacks-and-load-configuration)
-- [Binary hardening](#binary-hardening)
-- [ELF note tables](#elf-note-tables)
-- [ELF and Mach-O initializer functions](#elf-and-mach-o-initializer-functions)
-- [ELF unwind function ranges](#elf-unwind-function-ranges)
-- [Apple IPS crash reports](#apple-ips-crash-reports)
-- [Linux kernel modules](#linux-kernel-modules)
-- [Linux kernel build config](#linux-kernel-build-config)
-- [CPython pyc files](#cpython-pyc-files)
-- [Erlang BEAM files](#erlang-beam-files)
-- [SQLite snapshots](#sqlite-snapshots)
-- [Recorded Linux crashes](#recorded-linux-crashes)
-- [Offline ELF layout](#offline-elf-layout)
-- [Offline WASM](#offline-wasm)
-- [Offline APK signatures](#offline-apk-signatures)
-- [Offline PE resource inventory](#offline-pe-resource-inventory)
-- [Offline PE and ELF export tables](#offline-pe-and-elf-export-tables)
-- [EVM bytecode](#evm-bytecode)
-- [Embedded GPU code objects](#embedded-gpu-code-objects)
-
-## Release exposure
-
-Use `audit_release_exposure` (`rea audit-release-exposure PATH`) when the question is what static analysis of a shipped `.app`, install directory, ASAR, or executable recovers. Each check has a stable id, a status, the observation behind it, and fixed remediation. Unknown and unassessed checks stay in the report. The report does not establish that the application is secure.
-
-`text.credential_markers` reports a rule id and location only. Matched bytes, previews, lengths, and hashes of matched bytes are omitted.
-
-`native.hardening`, `macho.hardened_runtime`, and `android.manifest` are not assessed here. Their next probes are `inspect_binary_hardening`, `inspect_signature`, and `inspect_android_package`.
-
-See [release exposure](https://github.com/morluto/rea/blob/main/docs/release-exposure.md).
-
-## Go build metadata
-
-Use `inspect_go_binary` for an explicit local ELF, PE or Mach-O (universal
-files need `architecture`) when the
-question concerns the embedded compiler version, module/dependency versions,
-local replacements, build settings, or static Go string literals packed without
-NUL terminators. Pass `include_string_literals: true`
-(`--include-string-literals`, optional `max_string_literals` /
-`--max-string-literals`) to recover bounded inferred word-aligned `(ptr, len)`
-UTF-8 string slices from read-only data and page them through `go`
-`string_literals` views. No `open_binary`, Go installation or
-target execution is required. The CLI equivalent is `inspect-go-binary PATH`.
-Preserve artifact SHA-256, original file ranges, framed module bytes and every
-unparsed module-text line. A missing record is an unknown, not proof of a
-non-Go binary; observed metadata does not authenticate its claims, and static
-`(ptr, len)` string recovery excludes compiler-inlined stack/immediate strings
-without proving runtime reachability. Function,
-source and type recovery remain a separate native-analysis question.
-
-See the [Go binary guide](https://github.com/morluto/rea/blob/main/docs/guides/native/go-binaries.md).
-
-## Model files
-
-Optional `view` selects one page in the initial call. Batch metadata keys or
-tensor names and select only needed fields; do not infer model behavior from
-declared architecture or tensor types.
-
-Use `inspect_model_artifact` (`inspect-model-artifact PATH`) for local GGUF v2/v3
-and safetensors declarations without a model runtime. Header-only mode does not
-read weights or establish full-file identity. `hash: "full"` explicitly streams
-all bytes. Retained `model-artifact` views page `metadata`, `tensors` and derived
-`types`, with exact batch filters and selected fields. Arrays retain typed counts
-and eight-element previews; omitted values, unknown types and runtime behavior
-stay explicit. See [model files](https://github.com/morluto/rea/blob/main/docs/model-artifacts.md).
-
-## IL2CPP metadata
-
-Use `inspect_il2cpp_metadata` for an explicit local Unity `global-metadata.dat`
-(versions 24, 27, 29, 31 and 39) when the question concerns assembly images,
-names, member owners, string literals, namespaces or index ranges. No `open_binary`, Unity installation or target
-execution is required. The CLI equivalent is `inspect-il2cpp-metadata PATH`.
-MCP defaults to counts with complete retained Evidence. Query `sections`,
-`images`, `types`, `methods`, `fields`, `properties`, `events`, `parameters` or
-`string_literals` through `inspect_analysis_view`; use exact names,
-batch indices and selected fields. Row indices are scoped by artifact SHA-256
-and collection; page offsets count filtered matches. Parameters support batch
-`method_indices` to read one method's parameters directly. Historical reports
-without member collections retain unknown counts rather than empty inventories. Preserve
-artifact SHA-256 and index widths. Unrecognized magic may mean encryption and
-is unsupported, not decrypted.
-
-Optional matching `binary_path` enables static native addresses and type
-references. Query `method_addresses` with indices or addresses/states, preserving
-every shared-body method and both artifact identities. Without a linked binary,
-native counts remain unknown. Byte-pattern fallback is inferred; no runtime
-behavior or same-build pairing is proved.
-See [IL2CPP metadata](https://github.com/morluto/rea/blob/main/docs/il2cpp-metadata.md).
-
-## Rust dependency metadata
-
-For selected crates, use `view: {collection: "packages", names: [...],
-fields: ["name", "version", "source"]}` on the first call. Include `dependencies`
-explicitly for direct index references. Select `detail: "summary"` for counts and
-ranges; retained Rust package/edge views support later filters. Keep complete
-parent Evidence accessible; CLI compact delivery requires `--evidence_output`.
-
-Use `inspect_rust_binary` for an explicit local ELF, PE, or Mach-O (universal
-files need `architecture`) when
-the question concerns the cargo-auditable crate list embedded in `.dep-v0`
-(Mach-O: `__DATA,.dep-v0`). No `open_binary`, Rust installation, or target
-execution is required. The CLI equivalent is `inspect-rust-binary PATH`.
-Preserve artifact SHA-256, package order, dependency indices, and the section
-file range. A missing section is null metadata, not proof the binary has no
-dependencies. An omitted package kind stays null. A rustc commit hash is
-inferred from a read-only `/rustc/<40-hex>/` path and does not by itself
-identify the compiler. See [Rust binaries](https://github.com/morluto/rea/blob/main/docs/rust-binaries.md).
-
-## PE Authenticode certificate tables
-
-Use `inspect_pe_certificates` for an explicit local PE32/PE32+ image when the
-question concerns the Authenticode certificate table, WIN_CERTIFICATE types, or
-embedded PKCS#7 signer and X.509 identity fields. No `open_binary`, Windows SDK,
-or target execution is required. The CLI equivalent is
-`inspect-pe-certificates PATH`. Preserve artifact SHA-256, table file ranges,
-payload hashes, original validity-time spellings, and
-`subject_key_identifier_hex` when the signer uses that form. A missing
-`IMAGE_DIRECTORY_ENTRY_SECURITY` directory is observed absence of that table,
-not proof the image is unsigned in every catalog or overlay sense. Signatures,
-chains, timestamps, and Authenticode digest coverage remain unobserved. See
-[PE Authenticode certificate tables](https://github.com/morluto/rea/blob/main/docs/pe-certificates.md).
-
-## PE import and delay-load tables
-
-Use `inspect_pe_imports` for an explicit local PE32/PE32+ image when the
-question concerns imported DLL and function names, ordinals, delay-load
-descriptors, or the pefile import hash. No `open_binary`, Windows SDK, or
-target execution is required. The CLI equivalent is `inspect-pe-imports PATH`.
-Preserve artifact SHA-256, directory file ranges, original DLL spellings, and
-ILT/IAT RVAs. A missing `IMAGE_DIRECTORY_ENTRY_IMPORT` directory is observed
-absence of that table, not proof the image never imported APIs through another
-table or overlay. Bound snapshots, loaded IAT contents, and export-forward
-resolution remain unobserved. See
-[PE import and delay-load tables](https://github.com/morluto/rea/blob/main/docs/pe-imports.md).
-
-## Windows Shell Link shortcuts
-
-Use `inspect_windows_shortcut` for an explicit local `.lnk` file when the
-question concerns the recorded target path, arguments, working directory, icon
-location, volume serial, network share names, ExtraData blocks, or Tracker
-droid identifiers. No `open_binary`, Windows SDK, COM, or target execution is
-required. The CLI equivalent is `inspect-windows-shortcut PATH`. Preserve
-artifact SHA-256, FILETIME raw decimals, IDList type bytes, and ExtraData
-signatures. Unknown IDList items and ExtraData signatures remain raw with
-SHA-256; they are not guessed. Version-1 UUID MAC and timestamp values are
-labelled derived. A missing optional section is observed absence of that
-structure. See
-[Windows Shell Link shortcuts](https://github.com/morluto/rea/blob/main/docs/guides/native/windows-shortcut.md).
-
-## PE TLS callbacks and load configuration
-
-Preserve each callback's RVA, file offset and section alongside its VA and the
-size-covered declaration fields. A missing directory is observed absence of
-that declaration.
-
-Use `inspect_pe_load_config` for an explicit local PE32/PE32+ image when the
-question concerns code that runs before the entry point through TLS callbacks,
-or the security cookie, SafeSEH, CFG, EH-continuation and XFG declarations in
-the load configuration. No `open_binary`, Windows SDK, or target execution is
-required. The CLI equivalent is `inspect-pe-load-config PATH`. Preserve artifact
-SHA-256, directory and callback-array file ranges, and callback VAs as
-hexadecimal strings. Fields outside the declared `Size` are `null`. Guard-flag
-mitigations carry `evidence_kind: "inferred"`; enforcement remains unobserved.
-See
-[PE TLS callbacks and load configuration](https://github.com/morluto/rea/blob/main/docs/pe-load-config.md).
-
-## Binary hardening
-
-Each property reports `enabled`, `disabled`, `partial`, `unknown`, or
-`not_applicable`, with the header field or import behind it.
-
-Use `inspect_binary_hardening` for an explicit local ELF, PE, or Mach-O file
-when the question concerns PIE or ASLR, NX, RELRO, stack-protector or FORTIFY
-signs, control-flow flags, or stripped state. No `open_binary`, pwntools, or
-target execution is required. The CLI equivalent is
-`inspect-binary-hardening PATH`. Universal Mach-O files return one image per
-slice. A missing import is `unknown`, not proof the protection was omitted.
-Authenticode and Mach-O code-signature rows report presence, not validity.
-`hardened_runtime` stays unknown and names `inspect_signature`. See
-[binary hardening](https://github.com/morluto/rea/blob/main/docs/binary-hardening.md).
-
-## ELF note tables
-
-Retain each note's owner, type, descriptor SHA-256, and PT_NOTE/SHT_NOTE ranges.
-
-Use `inspect_elf_notes` for an explicit local ELF32/ELF64 image when the
-question concerns PT_NOTE/SHT_NOTE owners, GNU build-id, or GNU ABI-tag
-fields. No `open_binary` or target execution is required. The CLI equivalent
-is `inspect-elf-notes PATH`. Preserve artifact SHA-256, table file ranges,
-descriptor hashes, and original owner spellings. A missing table is observed
-absence of those declarations, not proof the image never carried notes in
-another container. See
-[ELF note tables](https://github.com/morluto/rea/blob/main/docs/elf-notes.md).
-
-## ELF and Mach-O initializer functions
-
-Use `inspect_native_initializers` for an explicit local ELF or Mach-O image
-when the question is which functions run before `main` or at exit: ELF
-`DT_PREINIT_ARRAY`, `DT_INIT`, `DT_INIT_ARRAY`, `DT_FINI`, `DT_FINI_ARRAY` and
-legacy `.ctors`/`.dtors`, or Mach-O `__mod_init_func`, `__init_offsets` and
-`__mod_term_func`. Pass `architecture` for a universal Mach-O. The CLI
-equivalent is `inspect-native-initializers PATH`. PIE slots are resolved
-through RELATIVE, RELR and section relocations, and Mach-O slots through
-chained fixups. Keep each entry's `resolution`: `unresolved` means the value was
-not established, and `unresolved_import` names an external symbol. Page with
-`offset` and `limit`. See
-[ELF and Mach-O initializer functions](https://github.com/morluto/rea/blob/main/docs/guides/native/native-initializers.md).
-
-## ELF unwind function ranges
-
-Use `inspect_elf_unwind_ranges` for an explicit local ELF32/ELF64 image when
-the question concerns function starts and lengths that survive `strip`, which
-functions have an LSDA, or which ranges have no exact symbol. No
-`open_binary` or target execution is required. The CLI equivalent is
-`inspect-elf-unwind-ranges PATH`. Preserve artifact SHA-256, FDE start and
-end, containing section, exact symbol name, CIE index, LSDA, and personality.
-Page later ranges with `offset` and `limit`. A missing `.eh_frame` is
-observed absence of that declaration and of `PT_GNU_EH_FRAME`, not proof the
-image has no functions. Header disagreement is a limitation and does not drop
-decoded ranges. `.debug_frame` is reported present or absent and is not
-decoded. CFA instructions are counted and not evaluated. Indirect EH pointers
-are not dereferenced. See
-[ELF unwind function ranges](https://github.com/morluto/rea/blob/main/docs/guides/native/elf-unwind-ranges.md).
-
-## Apple IPS crash reports
-
-Use `inspect_apple_ips` for an explicit local Apple `.ips` file when the
-question concerns crash metadata, exception and termination facts, threads,
-usedImages, lossless registers, or documented crash classes. No `open_binary`,
-Xcode, atos, or target execution is required. The CLI equivalent is
-`inspect-apple-ips PATH`. Preserve artifact SHA-256, `bug_type` as a string,
-uppercase image UUIDs, and integer literals that exceed IEEE-754 range. Only
-`bug_type` `"309"` yields a crash body; other DiagnosticReports types return
-no crash result. Select `join_path` for one local Mach-O or dSYM to match
-report images to slices by UUID and architecture. Preserve the selected
-artifact identity, slice facts and unjoined reasons; a joined static address
-does not resolve function names or dSYM symbols. See
-[Apple IPS crash reports](https://github.com/morluto/rea/blob/main/docs/apple-ips.md).
-
-## Linux kernel modules
-
-Use `inspect_kernel_module` for an explicit local ELF32/ELF64 relocatable
-object when the question concerns `.modinfo` license, vermagic, alias,
-depends, or parm strings, the `__versions` symbol CRCs, or an appended module
-signature trailer. No `open_binary`, Linux host, or `modinfo` program is
-required. The CLI equivalent is `inspect-kernel-module PATH`. Preserve
-artifact SHA-256, key order, duplicate keys, CRC hex, and section file ranges.
-A missing section is observed absence when section names are known, not proof
-the module never carried that metadata. `.gnu.linkonce.this_module` is reported
-as present or absent; `struct module` stays unknown because its layout changes
-with the kernel version. `__ksymtab` records are not decoded. The PKCS#7 range
-and its SHA-256 are observed; certificate, signer, and digest coverage remain
-unvalidated. An absent trailer is not proof the module is unsigned in another
-container. gzip, xz, and zstd whole-file containers are unsupported and name
-`external_decompression` as the next probe: use an external codec utility to
-decompress into a separate file, then call `inspect_kernel_module` on that file. See
-[Linux kernel modules](https://github.com/morluto/rea/blob/main/docs/kernel-modules.md).
-
-## Linux kernel build config
-
-Use `inspect_kernel_build_config` for an explicit local kernel image (`vmlinux`
-ELF, raw arm64 `Image`, x86 `bzImage`/`zImage`, or `configs.ko`) when the
-question is which kernel version it is or how it was configured. No
-`open_binary`, Linux host, or `extract-ikconfig` is required. The CLI
-equivalent is `inspect-kernel-build-config PATH`; pass `name` for one exact
-`CONFIG_*` option and `offset`/`limit` to page the rest. Preserve the banner
-version, builder, and compiler, the config SHA-256 and option count, and each
-entry's value kind (`y`, `m`, `string`, `number`, `not_set`). A config is
-accepted only when `IKCFG_ST` is immediately followed by a valid gzip member
-and `IKCFG_ED`, so an ordinary `IKCFG_ST` string is not a config. Offsets with
-`space: "inflated_payload"` are positions in an inflated bzImage payload, not
-file offsets. An absent config describes only the searched bytes; the kernel
-may have been built without `CONFIG_IKCONFIG` or with it as a module. Truncated
-payloads are incomplete, not absent. A framed blob that does not decode, has a
-bad CRC, or holds non-config text is `malformed`, not absent. xz, lz4, lzma, zstd, bzip2, and lzo
-payloads are unsupported and name `external_decompression`: decompress with an
-external utility, then call `inspect_kernel_build_config` on the result. See
-[Linux kernel build config](https://github.com/morluto/rea/blob/main/docs/guides/native/kernel-build-config.md).
-
-## CPython pyc files
-
-Use `inspect_python_pyc` for an explicit local CPython 3.7–3.13 `.pyc` when the
-question concerns the header magic, PEP 552 source hash or timestamp, nested
-code-object names, or bytecode identity. No `open_binary`, Python installation,
-or target execution is required. The CLI equivalent is `inspect-python-pyc PATH`.
-Preserve artifact SHA-256, the 16-byte header fields, nested code names, and
-bytecode SHA-256. Bytecode is not decompiled. A decoded header does not prove
-the matching `.py` is present. See [CPython pyc inspection](https://github.com/morluto/rea/blob/main/docs/python-pyc.md).
-
-## Erlang BEAM files
-
-Use `inspect_erlang_beam` for an explicit local `FOR1`/`BEAM` file when the
-question concerns IFF chunks, atom names, exports, imports, lambdas, or the
-Code header. No `open_binary`, Erlang installation, or bytecode execution is
-required. The CLI equivalent is `inspect-erlang-beam PATH`. Preserve artifact
-SHA-256, chunk file ranges, 1-based atom indices, and original atom spellings.
-Missing optional tables stay null. For a few facts, use `detail: "summary"` with
-a `beam` view selecting a collection, exact names/indices and fields; read later
-pages from `parent_evidence_id` or the complete Evidence bundle. Same-name
-exports retain every arity and original atom index. CLI summary delivery requires
-`--evidence_output`. Null table counts remain distinct from decoded empty tables. See
-[Erlang BEAM files](https://github.com/morluto/rea/blob/main/docs/erlang-beam.md).
-
-## SQLite snapshots
-
-Use `inspect_sqlite_database` with an explicit local database snapshot for its
-schema and optional rows from one exact ordinary table. It requires Node.js
-24.x >=24.15 or 26+ with the native `DatabaseSync.limits` API. It also consumes the
-existing sibling `-wal`; retain both original digests when citing its result.
-Default output is a compact inventory; a selected table expands only its own
-schema and samples 20 rows. Select `columns` to avoid irrelevant or large cells,
-and request `schema_scope: "database"` only for complete schema details.
-Choose `row_limit` when a finite sample is useful and keep truncation and
-unspecified row order explicit. Virtual tables, shadow tables and views are
-schema facts, not row-read targets. REA inspects private copies and leaves
-the source files untouched; obtain a coherent quiescent database/WAL snapshot
-before inspection. See [SQLite snapshots](https://github.com/morluto/rea/blob/main/docs/guides/data/sqlite-databases.md).
-
-## Recorded Linux crashes
-
-Use `inspect_recorded_crash` with an explicit Linux x86-64 ELF core path to
-inspect historical thread registers, signals, raw notes and original file
-ranges. It is independent of the active disassembler target. Source-bound
-register values and note bytes are observations; missing notes, signal-thread
-association and current executable/library identity remain unknown. PIDs are
-historical metadata and never authorize live attach or control.
-
-The optional `include_debugger_context` facet adds mapping candidates through
-caller-supplied GDB/pwndbg in an owned core-only session. Display names do not
-establish current file identity; zero reported flags leave permissions unknown.
-Requested unavailable context fails with setup guidance. See
-[recorded crashes](https://github.com/morluto/rea/blob/main/docs/guides/native/recorded-crashes.md)
-for exact upstream profiles, bounds and verification coverage.
-
-## Offline ELF layout
-
-Use `inspect_binary_layout` with an explicit local ELF path for file-backed
-sections, symbols, relocations, linkage names, and static mitigation inferences
-without opening a disassembler target. When that complete record is already
-retained, call `inspect_analysis_view` for a summary, mitigations or linkage
-facet, one section or symbol, or a stable page. Do not re-run the decoder to
-read a single object.
-
-## Offline WASM
-
-Supply WABT 1.0.42 through `REA_WABT_BIN_DIRECTORY`. Select JavaScript glue and
-candidate paths explicitly, preserving URL/path ambiguity and exact byte/tool
-identity; no implicit fetch is performed.
-
-Use `inspect_wasm_artifact` with caller-supplied WABT for selected local bytes.
-For counts, validation and WAT identity, select `detail: "summary"` on the first
-MCP call. The complete original Evidence stays retained; use
-`normalized_result.parent_evidence_id` for exact exports or text pages. Omitted
-or complete detail keeps the existing complete result.
-For a completed result, use `inspect_analysis_view` with its reported retained
-reference for a summary or `view: {kind: "wasm", facet: "exports", offset: 0, limit: 32}`.
-Facets also include sections, imports, candidates, glue, WAT, headers, details
-and tool profile. Text offsets count UTF-16 code units; follow `next_offset`.
-The complete source remains retained after an oversized MCP reply. No target
-execution or network acquisition is implied, and WAT is decoded representation.
-
-For exact core-module import/export differences, use `compare_wasm_interfaces`
-with two complete inspection Evidence records (or current-session retained
-references) and explicit scope. Preserve unknown types/duplicate names; static
-interface equality is not runtime compatibility. CLI accepts complete inline
-Evidence in `rea compare-wasm-interfaces input.json --json`, or exact references
-with `--evidence-bundle /absolute/path/evidence.json`. Use the complete parent
-ID when the inspection returned a summary.
-
-See the [WASM guide](https://github.com/morluto/rea/blob/main/docs/guides/wasm/wasm-artifacts.md).
-
-## Offline APK signatures
-
-Use `inspect_apk_signature` for a selected local APK/ZIP/AAB. Default MCP summaries
-retain complete Evidence. Select `view.collection: signers/digests/lineage`,
-with scheme or ID batches, fields and paging; use `inspect_analysis_view` kind
-`apk-signature` for retained reads after deletion. IDs retain artifact/scheme
-scope and signer links; `fields: []` keeps identity links. Empty pages preserve
-partial decoding reasons and unknown authenticity. CLI compact delivery needs
-`--evidence_output` to preserve the complete parent.
-It reads the APK
-Signing Block (v2/v3/v3.1) and v1 `META-INF` signing without executing anything,
-and reports each signer's certificate facts (for v1, the certificate the PKCS#7
-block's SignerInfo names), the declared v3 rotation lineage,
-the declared source-stamp certificate and, for SHA-256/SHA-512 digests, whether
-the declared content digest matches the current bytes. That match is not a
-signature verification: anyone who changes the bytes can recompute the digest.
-`posture` summarises v1-only signing (which cannot tell a v1-only APK from one
-whose v2/v3 block was removed), observed signer agreement across schemes,
-whether the v1 signer equals the block-scheme signer (`false` is notable),
-whether every differing signer appears in a declared (unverified) rotation
-lineage, and the strongest scheme.
-It never claims authenticity: no signature or lineage is cryptographically
-verified.
-
-See the [APK signature guide](https://github.com/morluto/rea/blob/main/docs/apk-signature.md).
-
-## Offline PE resource inventory
-
-Use `inspect_pe_resources` with an absolute `path`, or
-`rea inspect-pe-resources ./selected.exe --json`, for portable PE32/PE32+
-resource inspection without opening a native target. Preserve returned Evidence:
-type/name/language IDs and UTF-16 names, original directory/data/payload ranges,
-code pages, payload hashes and static icon-group references. Same-language icon
-resolution is explicit; missing references and size mismatch remain visible.
-Other payloads are opaque, and this operation does not decode images or simulate
-Windows resource loading. Consult the budgets and supported tree profile in
-[offline binary diagnostics](https://github.com/morluto/rea/blob/main/docs/guides/native/binary-diagnostics.md).
-
-## Offline PE and ELF export tables
-
-Use `inspect_native_exports` with an absolute `path`, or
-`rea inspect-native-exports ./libexample.so --json`, to read a PE export
-directory or ELF dynamic symbol table on any host without opening a native
-target. Ordinal-only, forwarded and hidden-version exports stay distinct;
-partial or unsupported coverage means the inventory is incomplete, not that
-exports are absent. Use `inspect_macho` for Mach-O. See
-[offline binary diagnostics](https://github.com/morluto/rea/blob/main/docs/guides/native/binary-diagnostics.md).
-
-## EVM bytecode
-
-Explicit local EVM bytecode carrier: `inspect_evm_interface` with caller-selected
-`raw` or `hex` encoding. The owned worker deadline is `timeout_ms` (default
-30000 ms, range 1..2147483647, next call). Select byte, address-space and CPU
-allowances separately through `producer_limits`; tighter inherited OS limits
-remain enforced, and the 256 MiB worker heap is independent. Preserve carrier/decoded digests and treat selectors,
-argument strings and mutability as inferred candidates; this performs no chain
-lookup or target execution. For the complete inferred interface without hex or
-duplicate stdout, discover support and select `detail: "summary"`; complete
-Evidence stays retained under its original identity. Use the returned
-`parent_evidence_id` for exact function or `bytecode_hex`/stdout/stderr pages.
-Candidate coverage does not establish interface completeness. See the
-[EVM bytecode guide](https://github.com/morluto/rea/blob/main/docs/guides/evm/evm-bytecode.md).
-
-## Embedded GPU code objects
-
-Embedded CUDA/HIP code objects: `inspect_gpu_code_objects`; inventory targets
-first, then decode selected payloads and query retained GPU Evidence.

.agents/skills/reverse-engineer-anything/references/javascript-applications.md

@@ -1,13 +1,5 @@
 # JavaScript and Electron application artifacts
 
-Read only the sections needed for the selected task.
-
-- [Inventory route candidates without returning the application graph](#inventory-route-candidates-without-returning-the-application-graph)
-- [Read a module without repeating analysis](#read-a-module-without-repeating-analysis)
-- [Trace only the remaining question](#trace-only-the-remaining-question)
-- [Recover an oversized display](#recover-an-oversized-display)
-- [Reusing application Evidence](#reusing-application-evidence)
-
 For captured or bundled JavaScript that needs readable source recovery, use
 `recover_javascript_sources` with an absolute `path` and absent
 `output_directory`. Supply related runtime/shared/lazy files in
@@ -98,9 +90,6 @@ Endpoint candidates remain inferences rather than confirmed network behavior.
 ## Read a module without repeating analysis
 
 Use the actual returned parent ID, on the same connection, as `source.evidence_id`.
-For CLI summary exports, take that ID from
-`summary.normalized_result.parent_evidence_id` and pass `--evidence-bundle` with
-the exported bundle's path to each later `inspect-analysis-view` invocation.
 The following IDs are placeholders to replace with returned values. When the
 module's node ID is not already known, page identities before selecting an item.
 Pages return only node IDs, paths and roles; choose a useful page size and follow

.agents/skills/reverse-engineer-anything/references/native-and-artifacts.md

@@ -1,23 +1,4 @@
-# Native and managed analysis
-
-<a id="native-managed-and-packaged-artifacts"></a>
-
-Read only the sections needed for the selected task.
-
-- [Other artifact guides](#other-artifact-guides)
-- [Apple app resources](#apple-app-resources)
-- [Desktop installation directories](#desktop-installation-directories)
-- [Native targets](#native-targets)
-- [Continue editable Ghidra analysis across processes](#continue-editable-ghidra-analysis-across-processes)
-- [Reuse caller-selected Ghidra analysis](#reuse-caller-selected-ghidra-analysis)
-- [Ghidra startup deadline recovery](#ghidra-startup-deadline-recovery)
-- [Bridge request ceiling](#bridge-request-ceiling)
-- [Selected native facts and call graphs](#selected-native-facts-and-call-graphs)
-- [Static memory and original file bytes](#static-memory-and-original-file-bytes)
-- [DOS MZ with Ghidra](#dos-mz-with-ghidra)
-- [Managed PE/CLI](#managed-pe-cli-guide)
-- [DOS COM loader interpretation](#dos-com-loader-interpretation)
-- [Function annotations in Ghidra](#function-annotations-in-ghidra)
+# Native, managed, and packaged artifacts
 
 For release comparisons on one MCP connection, retain `list_procedures` Evidence
 from each target and reuse its exact IDs after target switches or `close_binary`.
@@ -81,39 +62,342 @@ runtime presence. For compact CLI output, use
 `inspect-desktop-application PATH --detail summary --evidence_output BUNDLE`.
 See [desktop installations](https://github.com/morluto/rea/blob/main/docs/desktop-applications.md).
 
+## Release exposure
+
+Use `audit_release_exposure` (`rea audit-release-exposure PATH`) when the question is what static analysis of a shipped `.app`, install directory, ASAR, or executable recovers. Each check has a stable id, a status, the observation behind it, and fixed remediation. Unknown and unassessed checks stay in the report. The report does not establish that the application is secure.
+
+`text.credential_markers` reports a rule id and location only. Matched bytes, previews, lengths, and hashes of matched bytes are omitted.
+
+`native.hardening`, `macho.hardened_runtime`, and `android.manifest` are not assessed here. Their next probes are `inspect_binary_hardening`, `inspect_signature`, and `inspect_android_package`.
+
+See [release exposure](https://github.com/morluto/rea/blob/main/docs/release-exposure.md).
+
+## Go build metadata
+
+Use `inspect_go_binary` for an explicit local ELF, PE or Mach-O (universal
+files need `architecture`) when the
+question concerns the embedded compiler version, module/dependency versions,
+local replacements, build settings, or static Go string literals packed without
+NUL terminators. Pass `include_string_literals: true`
+(`--include-string-literals`, optional `max_string_literals` /
+`--max-string-literals`) to recover bounded inferred word-aligned `(ptr, len)`
+UTF-8 string slices from read-only data and page them through `go`
+`string_literals` views. No `open_binary`, Go installation or
+target execution is required. The CLI equivalent is `inspect-go-binary PATH`.
+Preserve artifact SHA-256, original file ranges, framed module bytes and every
+unparsed module-text line. A missing record is an unknown, not proof of a
+non-Go binary; observed metadata does not authenticate its claims, and static
+`(ptr, len)` string recovery excludes compiler-inlined stack/immediate strings
+without proving runtime reachability. Function,
+source and type recovery remain a separate native-analysis question.
+
+## Model files
+
+Use `inspect_model_artifact` (`inspect-model-artifact PATH`) for local GGUF v2/v3
+and safetensors declarations without a model runtime. Header-only mode does not
+read weights or establish full-file identity. `hash: "full"` explicitly streams
+all bytes. Retained `model-artifact` views page `metadata`, `tensors` and derived
+`types`, with exact batch filters and selected fields. Arrays retain typed counts
+and eight-element previews; omitted values, unknown types and runtime behavior
+stay explicit. See [model files](https://github.com/morluto/rea/blob/main/docs/model-artifacts.md).
+
+## IL2CPP metadata
+
+Use `inspect_il2cpp_metadata` for an explicit local Unity `global-metadata.dat`
+(versions 24, 27, 29, 31 and 39) when the question concerns assembly images,
+names, member owners, string literals, namespaces or index ranges. No `open_binary`, Unity installation or target
+execution is required. The CLI equivalent is `inspect-il2cpp-metadata PATH`.
+MCP defaults to counts with complete retained Evidence. Query `sections`,
+`images`, `types`, `methods`, `fields`, `properties`, `events`, `parameters` or
+`string_literals` through `inspect_analysis_view`; use exact names,
+batch indices and selected fields. Row indices are scoped by artifact SHA-256
+and collection; page offsets count filtered matches. Parameters support batch
+`method_indices` to read one method's parameters directly. Historical reports
+without member collections retain unknown counts rather than empty inventories. Preserve
+artifact SHA-256 and index widths. Unrecognized magic may mean encryption and
+is unsupported, not decrypted. Linking names to `GameAssembly` or
+`UnityFramework` code addresses remains a separate native-analysis question.
+See [IL2CPP metadata](https://github.com/morluto/rea/blob/main/docs/il2cpp-metadata.md).
+
+## Rust dependency metadata
+
+Use `inspect_rust_binary` for an explicit local ELF, PE, or Mach-O (universal
+files need `architecture`) when
+the question concerns the cargo-auditable crate list embedded in `.dep-v0`
+(Mach-O: `__DATA,.dep-v0`). No `open_binary`, Rust installation, or target
+execution is required. The CLI equivalent is `inspect-rust-binary PATH`.
+Preserve artifact SHA-256, package order, dependency indices, and the section
+file range. A missing section is null metadata, not proof the binary has no
+dependencies. An omitted package kind stays null. A rustc commit hash is
+inferred from a read-only `/rustc/<40-hex>/` path and does not by itself
+identify the compiler. See [Rust binaries](https://github.com/morluto/rea/blob/main/docs/rust-binaries.md).
+
+## PE Authenticode certificate tables
+
+Use `inspect_pe_certificates` for an explicit local PE32/PE32+ image when the
+question concerns the Authenticode certificate table, WIN_CERTIFICATE types, or
+embedded PKCS#7 signer and X.509 identity fields. No `open_binary`, Windows SDK,
+or target execution is required. The CLI equivalent is
+`inspect-pe-certificates PATH`. Preserve artifact SHA-256, table file ranges,
+payload hashes, original validity-time spellings, and
+`subject_key_identifier_hex` when the signer uses that form. A missing
+`IMAGE_DIRECTORY_ENTRY_SECURITY` directory is observed absence of that table,
+not proof the image is unsigned in every catalog or overlay sense. Signatures,
+chains, timestamps, and Authenticode digest coverage remain unobserved. See
+[PE Authenticode certificate tables](https://github.com/morluto/rea/blob/main/docs/pe-certificates.md).
+
+## PE import and delay-load tables
+
+Use `inspect_pe_imports` for an explicit local PE32/PE32+ image when the
+question concerns imported DLL and function names, ordinals, delay-load
+descriptors, or the pefile import hash. No `open_binary`, Windows SDK, or
+target execution is required. The CLI equivalent is `inspect-pe-imports PATH`.
+Preserve artifact SHA-256, directory file ranges, original DLL spellings, and
+ILT/IAT RVAs. A missing `IMAGE_DIRECTORY_ENTRY_IMPORT` directory is observed
+absence of that table, not proof the image never imported APIs through another
+table or overlay. Bound snapshots, loaded IAT contents, and export-forward
+resolution remain unobserved. See
+[PE import and delay-load tables](https://github.com/morluto/rea/blob/main/docs/pe-imports.md).
+
+## Windows Shell Link shortcuts
+
+Use `inspect_windows_shortcut` for an explicit local `.lnk` file when the
+question concerns the recorded target path, arguments, working directory, icon
+location, volume serial, network share names, ExtraData blocks, or Tracker
+droid identifiers. No `open_binary`, Windows SDK, COM, or target execution is
+required. The CLI equivalent is `inspect-windows-shortcut PATH`. Preserve
+artifact SHA-256, FILETIME raw decimals, IDList type bytes, and ExtraData
+signatures. Unknown IDList items and ExtraData signatures remain raw with
+SHA-256; they are not guessed. Version-1 UUID MAC and timestamp values are
+labelled derived. A missing optional section is observed absence of that
+structure. See
+[Windows Shell Link shortcuts](https://github.com/morluto/rea/blob/main/docs/guides/native/windows-shortcut.md).
+
+## PE TLS callbacks and load configuration
+
+Use `inspect_pe_load_config` for an explicit local PE32/PE32+ image when the
+question concerns code that runs before the entry point through TLS callbacks,
+or the security cookie, SafeSEH, CFG, EH-continuation and XFG declarations in
+the load configuration. No `open_binary`, Windows SDK, or target execution is
+required. The CLI equivalent is `inspect-pe-load-config PATH`. Preserve artifact
+SHA-256, directory and callback-array file ranges, and callback VAs as
+hexadecimal strings. Fields outside the declared `Size` are `null`. Guard-flag
+mitigations carry `evidence_kind: "inferred"`; enforcement remains unobserved.
+See
+[PE TLS callbacks and load configuration](https://github.com/morluto/rea/blob/main/docs/pe-load-config.md).
+
+## Binary hardening
+
+Use `inspect_binary_hardening` for an explicit local ELF, PE, or Mach-O file
+when the question concerns PIE or ASLR, NX, RELRO, stack-protector or FORTIFY
+signs, control-flow flags, or stripped state. No `open_binary`, pwntools, or
+target execution is required. The CLI equivalent is
+`inspect-binary-hardening PATH`. Universal Mach-O files return one image per
+slice. A missing import is `unknown`, not proof the protection was omitted.
+Authenticode and Mach-O code-signature rows report presence, not validity.
+`hardened_runtime` stays unknown and names `inspect_signature`. See
+[binary hardening](https://github.com/morluto/rea/blob/main/docs/binary-hardening.md).
+
+## ELF note tables
+
+Use `inspect_elf_notes` for an explicit local ELF32/ELF64 image when the
+question concerns PT_NOTE/SHT_NOTE owners, GNU build-id, or GNU ABI-tag
+fields. No `open_binary` or target execution is required. The CLI equivalent
+is `inspect-elf-notes PATH`. Preserve artifact SHA-256, table file ranges,
+descriptor hashes, and original owner spellings. A missing table is observed
+absence of those declarations, not proof the image never carried notes in
+another container. See
+[ELF note tables](https://github.com/morluto/rea/blob/main/docs/elf-notes.md).
+
+## ELF and Mach-O initializer functions
+
+Use `inspect_native_initializers` for an explicit local ELF or Mach-O image
+when the question is which functions run before `main` or at exit: ELF
+`DT_PREINIT_ARRAY`, `DT_INIT`, `DT_INIT_ARRAY`, `DT_FINI`, `DT_FINI_ARRAY` and
+legacy `.ctors`/`.dtors`, or Mach-O `__mod_init_func`, `__init_offsets` and
+`__mod_term_func`. Pass `architecture` for a universal Mach-O. The CLI
+equivalent is `inspect-native-initializers PATH`. PIE slots are resolved
+through RELATIVE, RELR and section relocations, and Mach-O slots through
+chained fixups. Keep each entry's `resolution`: `unresolved` means the value was
+not established, and `unresolved_import` names an external symbol. Page with
+`offset` and `limit`. See
+[ELF and Mach-O initializer functions](https://github.com/morluto/rea/blob/main/docs/guides/native/native-initializers.md).
+
+## ELF unwind function ranges
+
+Use `inspect_elf_unwind_ranges` for an explicit local ELF32/ELF64 image when
+the question concerns function starts and lengths that survive `strip`, which
+functions have an LSDA, or which ranges have no exact symbol. No
+`open_binary` or target execution is required. The CLI equivalent is
+`inspect-elf-unwind-ranges PATH`. Preserve artifact SHA-256, FDE start and
+end, containing section, exact symbol name, CIE index, LSDA, and personality.
+Page later ranges with `offset` and `limit`. A missing `.eh_frame` is
+observed absence of that declaration and of `PT_GNU_EH_FRAME`, not proof the
+image has no functions. Header disagreement is a limitation and does not drop
+decoded ranges. `.debug_frame` is reported present or absent and is not
+decoded. CFA instructions are counted and not evaluated. Indirect EH pointers
+are not dereferenced. See
+[ELF unwind function ranges](https://github.com/morluto/rea/blob/main/docs/guides/native/elf-unwind-ranges.md).
+
+## Apple IPS crash reports
+
+Use `inspect_apple_ips` for an explicit local Apple `.ips` file when the
+question concerns crash metadata, exception and termination facts, threads,
+usedImages, lossless registers, or documented crash classes. No `open_binary`,
+Xcode, atos, or target execution is required. The CLI equivalent is
+`inspect-apple-ips PATH`. Preserve artifact SHA-256, `bug_type` as a string,
+uppercase image UUIDs, and integer literals that exceed IEEE-754 range. Only
+`bug_type` `"309"` yields a crash body; other DiagnosticReports types return
+no crash result. Select `join_path` for one local Mach-O or dSYM to match
+report images to slices by UUID and architecture. Preserve the selected
+artifact identity, slice facts and unjoined reasons; a joined static address
+does not resolve function names or dSYM symbols. See
+[Apple IPS crash reports](https://github.com/morluto/rea/blob/main/docs/apple-ips.md).
+
+## Linux kernel modules
+
+Use `inspect_kernel_module` for an explicit local ELF32/ELF64 relocatable
+object when the question concerns `.modinfo` license, vermagic, alias,
+depends, or parm strings, the `__versions` symbol CRCs, or an appended module
+signature trailer. No `open_binary`, Linux host, or `modinfo` program is
+required. The CLI equivalent is `inspect-kernel-module PATH`. Preserve
+artifact SHA-256, key order, duplicate keys, CRC hex, and section file ranges.
+A missing section is observed absence when section names are known, not proof
+the module never carried that metadata. `.gnu.linkonce.this_module` is reported
+as present or absent; `struct module` stays unknown because its layout changes
+with the kernel version. `__ksymtab` records are not decoded. The PKCS#7 range
+and its SHA-256 are observed; certificate, signer, and digest coverage remain
+unvalidated. An absent trailer is not proof the module is unsigned in another
+container. gzip, xz, and zstd whole-file containers are unsupported and name
+`external_decompression` as the next probe: use an external codec utility to
+decompress into a separate file, then call `inspect_kernel_module` on that file. See
+[Linux kernel modules](https://github.com/morluto/rea/blob/main/docs/kernel-modules.md).
+
+## Linux kernel build config
+
+Use `inspect_kernel_build_config` for an explicit local kernel image (`vmlinux`
+ELF, raw arm64 `Image`, x86 `bzImage`/`zImage`, or `configs.ko`) when the
+question is which kernel version it is or how it was configured. No
+`open_binary`, Linux host, or `extract-ikconfig` is required. The CLI
+equivalent is `inspect-kernel-build-config PATH`; pass `name` for one exact
+`CONFIG_*` option and `offset`/`limit` to page the rest. Preserve the banner
+version, builder, and compiler, the config SHA-256 and option count, and each
+entry's value kind (`y`, `m`, `string`, `number`, `not_set`). A config is
+accepted only when `IKCFG_ST` is immediately followed by a valid gzip member
+and `IKCFG_ED`, so an ordinary `IKCFG_ST` string is not a config. Offsets with
+`space: "inflated_payload"` are positions in an inflated bzImage payload, not
+file offsets. An absent config describes only the searched bytes; the kernel
+may have been built without `CONFIG_IKCONFIG` or with it as a module. Truncated
+payloads are incomplete, not absent. A framed blob that does not decode, has a
+bad CRC, or holds non-config text is `malformed`, not absent. xz, lz4, lzma, zstd, bzip2, and lzo
+payloads are unsupported and name `external_decompression`: decompress with an
+external utility, then call `inspect_kernel_build_config` on the result. See
+[Linux kernel build config](https://github.com/morluto/rea/blob/main/docs/guides/native/kernel-build-config.md).
+
+## CPython pyc files
+
+Use `inspect_python_pyc` for an explicit local CPython 3.7–3.13 `.pyc` when the
+question concerns the header magic, PEP 552 source hash or timestamp, nested
+code-object names, or bytecode identity. No `open_binary`, Python installation,
+or target execution is required. The CLI equivalent is `inspect-python-pyc PATH`.
+Preserve artifact SHA-256, the 16-byte header fields, nested code names, and
+bytecode SHA-256. Bytecode is not decompiled. A decoded header does not prove
+the matching `.py` is present. See [CPython pyc inspection](https://github.com/morluto/rea/blob/main/docs/python-pyc.md).
+
+## Erlang BEAM files
+
+Use `inspect_erlang_beam` for an explicit local `FOR1`/`BEAM` file when the
+question concerns IFF chunks, atom names, exports, imports, lambdas, or the
+Code header. No `open_binary`, Erlang installation, or bytecode execution is
+required. The CLI equivalent is `inspect-erlang-beam PATH`. Preserve artifact
+SHA-256, chunk file ranges, 1-based atom indices, and original atom spellings.
+Missing optional tables stay null. For a few facts, use `detail: "summary"` with
+a `beam` view selecting a collection, exact names/indices and fields; read later
+pages from `parent_evidence_id` or the complete Evidence bundle. Same-name
+exports retain every arity and original atom index. CLI summary delivery requires
+`--evidence_output`. Null table counts remain distinct from decoded empty tables. See
+[Erlang BEAM files](https://github.com/morluto/rea/blob/main/docs/erlang-beam.md).
+
+## SQLite snapshots
+
+Use `inspect_sqlite_database` with an explicit local database snapshot for its
+schema and optional rows from one exact ordinary table. It requires Node.js
+24.x >=24.15 or 26+ with the native `DatabaseSync.limits` API. It also consumes the
+existing sibling `-wal`; retain both original digests when citing its result.
+Default output is a compact inventory; a selected table expands only its own
+schema and samples 20 rows. Select `columns` to avoid irrelevant or large cells,
+and request `schema_scope: "database"` only for complete schema details.
+Choose `row_limit` when a finite sample is useful and keep truncation and
+unspecified row order explicit. Virtual tables, shadow tables and views are
+schema facts, not row-read targets. REA inspects private copies and leaves
+the source files untouched; obtain a coherent quiescent database/WAL snapshot
+before inspection. See [SQLite snapshots](https://github.com/morluto/rea/blob/main/docs/guides/data/sqlite-databases.md).
+
+## Recorded Linux crashes
+
+Use `inspect_recorded_crash` with an explicit Linux x86-64 ELF core path to
+inspect historical thread registers, signals, raw notes and original file
+ranges. It is independent of the active disassembler target. Source-bound
+register values and note bytes are observations; missing notes, signal-thread
+association and current executable/library identity remain unknown. PIDs are
+historical metadata and never authorize live attach or control.
+
+The optional `include_debugger_context` facet adds mapping candidates through
+caller-supplied GDB/pwndbg in an owned core-only session. Display names do not
+establish current file identity; zero reported flags leave permissions unknown.
+Requested unavailable context fails with setup guidance. See
+[recorded crashes](https://github.com/morluto/rea/blob/main/docs/guides/native/recorded-crashes.md)
+for exact upstream profiles, bounds and verification coverage.
+
+## Offline ELF layout
+
+Use `inspect_binary_layout` with an explicit local ELF path for file-backed
+sections, symbols, relocations, linkage names, and static mitigation inferences
+without opening a disassembler target. When that complete record is already
+retained, call `inspect_analysis_view` for a summary, mitigations or linkage
+facet, one section or symbol, or a stable page. Do not re-run the decoder to
+read a single object.
+
+## Offline WASM
+
+Use `inspect_wasm_artifact` with caller-supplied WABT for selected local bytes.
+For counts, validation and WAT identity, select `detail: "summary"` on the first
+MCP call. The complete original Evidence stays retained; use
+`normalized_result.parent_evidence_id` for exact exports or text pages. Omitted
+or complete detail keeps the existing complete result.
+For a completed result, use `inspect_analysis_view` with its reported retained
+reference for a summary or `view: {kind: "wasm", facet: "exports", offset: 0, limit: 32}`.
+Facets also include sections, imports, candidates, glue, WAT, headers, details
+and tool profile. Text offsets count UTF-16 code units; follow `next_offset`.
+The complete source remains retained after an oversized MCP reply. No target
+execution or network acquisition is implied, and WAT is decoded representation.
+
+## Offline APK signatures
+
+Use `inspect_apk_signature` for a selected local APK/ZIP/AAB. Default MCP summaries
+retain complete Evidence. Select `view.collection: signers/digests/lineage`,
+with scheme or ID batches, fields and paging; use `inspect_analysis_view` kind
+`apk-signature` for retained reads after deletion. IDs retain artifact/scheme
+scope and signer links; `fields: []` keeps identity links. Empty pages preserve
+partial decoding reasons and unknown authenticity. CLI compact delivery needs
+`--evidence_output` to preserve the complete parent.
+It reads the APK
+Signing Block (v2/v3/v3.1) and v1 `META-INF` signing without executing anything,
+and reports each signer's certificate facts (for v1, the certificate the PKCS#7
+block's SignerInfo names), the declared v3 rotation lineage,
+the declared source-stamp certificate and, for SHA-256/SHA-512 digests, whether
+the declared content digest matches the current bytes. That match is not a
+signature verification: anyone who changes the bytes can recompute the digest.
+`posture` summarises v1-only signing (which cannot tell a v1-only APK from one
+whose v2/v3 block was removed), observed signer agreement across schemes,
+whether the v1 signer equals the block-scheme signer (`false` is notable),
+whether every differing signer appears in a declared (unverified) rotation
+lineage, and the strongest scheme.
+It never claims authenticity: no signature or lineage is cryptographically
+verified.
+
 ## Native targets
 
-For an existing IDA MCP configuration, use `open_binary` with the original
-input binary and `provider_id: "ida"`; do not pass an `.idb` or `.i64` database.
-The legacy attached profile binds the already-open GUI input by SHA-256 and
-leaves its database open. The modern headless profile creates and closes an
-owned private database without saving. Installation and registration reuse
-[mrexodia/ida-pro-mcp](https://github.com/mrexodia/ida-pro-mcp); see the
-[IDA provider guide](https://github.com/morluto/rea/blob/main/docs/providers/ida-provider.md).
-For migration to the separate native IDA MCP plugin, follow the guide's
-[legacy attached migration](https://github.com/morluto/rea/blob/main/docs/providers/ida-provider.md#migrate-a-legacy-attached-registration).
-Replace the upstream registration explicitly; modern mrexodia tools are not the
-native profile. Native attachment preserves the database but does not provide
-direct callers; check session availability and the guide's verification limits.
-Proceed directly to `analyze_function`, pseudocode, or function/string searches
-after opening; IDA does not supply `binary_overview`. Consult session availability
-when composing broader workflows. Modern direct callers and unsupported dossier
-facets remain unknown; live IDA results are not replayed from snapshots.
-
-With `REA_KUNA_COMMAND` configured, `provider_id: "kuna"` answers procedure
-lists, lookups, pseudocode, assembly and `analyze_function` from Kuna's cached
-whole-program export. Callers/callees use a separate body-free graph shared by
-the session; `xrefs` returns exact-address incoming sites and filters forwarding
-aliases. These queries and inventory lookups do not trigger a C export. Indirect
-targets remain unknown and can make relationship coverage partial. Reference
-caches expire on close; a fresh CLI process repeats the native reference walk.
-Typed references, CFG, raw-memory tools and annotations remain unsupported. A cold first
-answer can come from a single-function run (its limitations say so); assembly
-waits for the export. For pseudocode and assembly, `raw_result.source` names the
-cached `.c`/`.asm` file and byte span, which you may search directly instead of
-calling per function. See the
-[Kuna provider guide](https://github.com/morluto/rea/blob/main/docs/kuna-provider.md).
-
 After `open_binary`, use focused search, procedure, or function tools directly.
 Use `binary_overview` when target metadata or inventory context helps answer the
 question. Prefer literal search, names, decompilation, callers, callees, and
@@ -348,8 +632,6 @@ that a class is a root class or has no ivars.
 
 ## Managed PE/CLI
 
-<a id="managed-pe-cli-guide"></a>
-
 Start with `inspect_managed_artifact`. REA's canonical managed inspection is
 execution-free: do not claim it loaded, reflected, executed, or resolved the
 assembly. Keep managed/native boundaries and unavailable reconstruction facts
@@ -408,17 +690,81 @@ and distinguish derived metadata from Ghidra database observations. Recovered
 types and native pseudocode are not original C# source. Windows Ghidra P0 has
 no database mutation authority.
 
-For headerless firmware and address-bearing containers, read
-[raw images](firmware-and-raw-images.md#headerless-raw-images-and-firmware-containers).
+## Packages and extraction
 
-## DOS COM loader interpretation
+Use `inspect_artifact` for application bundles, archives, ZIP/APK/IPA/MSIX/AppX,
+ASAR, or DMG inputs when the artifact graph and findings help answer the
+question. It returns the complete artifact graph inline. Cite graph manifest
+IDs when using them.
+
+For an IPA or a macOS `.app`, ZIP, or DMG, pass the inventory Evidence to
+`project_apple_application_graph`. It reports application roots and nested
+bundles by path convention: app extensions, XPC services, frameworks, login
+items, system and driver extensions, and plug-ins. It also lists privileged
+helpers, launchd plists, symlinks, and each bundle's `info_plist_path` and
+executable candidates. Roles are path conventions, not parsed plists; read the
+listed plists with `inspect_plist`.
+
+After `open_binary` on a `.app` or Mach-O, use `trace_dylib_resolution` to see
+which file each `@rpath`, `@loader_path` and `@executable_path` load reaches for
+every executable in the bundle. It also lists missing or weak loads and earlier
+`@rpath` candidates that are absent. Narrow large bundles with `roots` or
+`architecture`. System paths stay undetermined because the dyld shared cache
+provides them.
+
+Use `extract_artifact` when materialized files from a supported format are needed.
+It extracts the active target and accepts optional `integrity_policy`; see the [MCP contracts](https://github.com/morluto/rea/blob/main/docs/reference/mcp-contracts.md).
+It materializes all regular files into a fresh temporary directory chosen by
+REA. Symlinks and encrypted entries are inventory facts, not extractable files.
+The result reports the materialized directory; callers do not choose the
+destination. No separate permission grant is needed.
+
+For classic iNES cartridges, `open_binary` recognizes the byte signature without
+an executable architecture. `inspect_artifact` retains header facts and each
+PRG/CHR file unit's offset, length and SHA-256; its nested inventory's
+`ines.reconstruction` verifies the actual ordered region bytes against the
+whole-file digest. Extraction includes the original header, optional trainer
+and uninterpreted trailing bytes. These file units are not mapper/CPU bank
+maps. NES 2.0, UNIF, dirty legacy headers and truncated payloads are explicitly
+refused. No emulator or 6502 analysis is performed. See the
+[iNES guide](https://github.com/morluto/rea/blob/main/docs/ines-containers.md).
+
+Native DMG traversal automatically uses a read-only mount on supported macOS
+hosts, with an owned temporary mount directory and cleanup. It needs no approval
+flag. Unsupported hosts or mount failures remain explicit limitations; do not
+claim child inventory when only root identity is available. DMG extraction is
+unavailable through `extract_artifact`. To access DMG files on macOS, create a
+read-only mount with your own tools, select files from that mount for analysis,
+and detach the mount you created when finished. REA's inspection mount is
+temporary, not a retained export of the image's contents.
 
 For headerless DOS COM, explicitly open with `format: "dos-com"` (CLI
 `--target-format dos-com`). Inspect the load image before trusting function
 analysis: the entry is `0x10100`, file offset zero, with imposed real-mode segment
 context. PSP/stack/device state remains unmodeled. See [DOS guide](https://github.com/morluto/rea/blob/main/docs/providers/ghidra-dos.md).
 
-## Function annotations in Ghidra
+Use `inspect_byte_entropy` to triage high- and low-entropy ranges before extraction.
+
+For other headerless images (embedded firmware without ELF/Mach-O/PE/MZ
+headers), open with `format: "raw-binary"`, `architecture` (`x86`, `x86_64`,
+`arm`, `arm64`, `6502`, or `m68k`), and `entry_point` (`0x` hex). File offset 0 is mapped at
+that address. Hopper and IDA refuse this interpretation; Ghidra uses
+`BinaryLoader` with a generic language: big-endian for `m68k`, little-endian
+otherwise. Big-endian ARM, Thumb-only, and Cortex-M variants are unsupported.
+`6502` selects NMOS 6502; the entry and complete file mapping must fit within
+`0x0000..0xffff`. 65C02, 65816, cartridge containers, bank switching, and mapper
+semantics are unsupported. `m68k` requires an even entry and decodes the 68040
+superset of the 68000 family; Amiga/Atari containers and vector-table entry
+discovery are unsupported. Windows Ghidra P0 admits raw images only for `6502`. See [raw binary guide](https://github.com/morluto/rea/blob/main/docs/providers/raw-binary.md).
+
+Intel HEX, S-record, UF2 and DfuSe files are containers, not raw images:
+`inspect_artifact` lists address regions (UF2 per familyID, DfuSe per element
+with its alternate setting), and `extract_artifact` writes each region. Open an
+extracted region as `raw-binary` with its start address as `entry_point`. UF2
+numbering gaps leave coverage unknown; a DfuSe suffix CRC mismatch follows
+`integrity_policy`. See [UF2 and DfuSe](https://github.com/morluto/rea/blob/main/docs/uf2-dfuse.md).
+
+### Function annotations in Ghidra
 
 Fresh-import sessions admit annotations; existing-project reuse is read-only.
 Use `annotate_native_function` for one function name and/or entry comments, with at least one explicit change. Review its annotation readback and refreshed dossier inline. Changes are atomic; empty comments clear them, omitted fields preserve them. Later MCP calls observe edits. Normal close saves them when owned Program checkpoint mode is enabled; ephemeral mode discards them. CLI `annotate-native-function` follows the same setting when closing after its updated result. Select the [checkpoint workflow](#continue-editable-ghidra-analysis-across-processes) before edits that must survive a later process. Original executable bytes are unchanged; edits invalidate immutable snapshots. Windows P0 does not admit database mutations. For naming only, `set_address_name` / `set_addresses_names` also work with Ghidra: they name function entries or data addresses (primary labels) and return compact per-address success booleans instead of a dossier. Ghidra rejects batch addresses that request the same name or a name another selected address holds; `raw_result.failures` gives each failure's reason.
@@ -438,39 +784,160 @@ architecture, provider-owned mapping and selected-image commitments. Equal
 numeric addresses alone cannot establish correspondence; inspect retained
 source Evidence and coverage when the match remains ambiguous.
 
-## Other artifact guides
+## Offline PE resource inventory
 
-Read the owning page for offline formats, containers or firmware. These explicit
-anchors preserve older links without creating duplicate searchable sections.
+Use `inspect_pe_resources` with an absolute `path`, or
+`rea inspect-pe-resources ./selected.exe --json`, for portable PE32/PE32+
+resource inspection without opening a native target. Preserve returned Evidence:
+type/name/language IDs and UTF-16 names, original directory/data/payload ranges,
+code pages, payload hashes and static icon-group references. Same-language icon
+resolution is explicit; missing references and size mismatch remain visible.
+Other payloads are opaque, and this operation does not decode images or simulate
+Windows resource loading. Consult the budgets and supported tree profile in
+[offline binary diagnostics](https://github.com/morluto/rea/blob/main/docs/guides/native/binary-diagnostics.md).
 
-- <a id="go-build-metadata"></a>[Go build metadata](format-inspectors.md#go-build-metadata)
-- <a id="model-files"></a>[Model files](format-inspectors.md#model-files)
-- <a id="il2cpp-metadata"></a>[IL2CPP metadata](format-inspectors.md#il2cpp-metadata)
-- <a id="rust-dependency-metadata"></a>[Rust dependency metadata](format-inspectors.md#rust-dependency-metadata)
-- <a id="pe-authenticode-certificate-tables"></a>[PE Authenticode certificate tables](format-inspectors.md#pe-authenticode-certificate-tables)
-- <a id="pe-import-and-delay-load-tables"></a>[PE import and delay-load tables](format-inspectors.md#pe-import-and-delay-load-tables)
-- <a id="pe-tls-callbacks-and-load-configuration"></a>[PE TLS callbacks and load configuration](format-inspectors.md#pe-tls-callbacks-and-load-configuration)
-- <a id="binary-hardening"></a>[Binary hardening](format-inspectors.md#binary-hardening)
-- <a id="elf-note-tables"></a>[ELF note tables](format-inspectors.md#elf-note-tables)
-- <a id="apple-ips-crash-reports"></a>[Apple IPS crash reports](format-inspectors.md#apple-ips-crash-reports)
-- <a id="linux-kernel-modules"></a>[Linux kernel modules](format-inspectors.md#linux-kernel-modules)
-- <a id="cpython-pyc-files"></a>[CPython pyc files](format-inspectors.md#cpython-pyc-files)
-- <a id="erlang-beam-files"></a>[Erlang BEAM files](format-inspectors.md#erlang-beam-files)
-- <a id="sqlite-snapshots"></a>[SQLite snapshots](format-inspectors.md#sqlite-snapshots)
-- <a id="recorded-linux-crashes"></a>[Recorded Linux crashes](format-inspectors.md#recorded-linux-crashes)
-- <a id="offline-elf-layout"></a>[Offline ELF layout](format-inspectors.md#offline-elf-layout)
-- <a id="offline-wasm"></a>[Offline WASM](format-inspectors.md#offline-wasm)
-- <a id="offline-apk-signatures"></a>[Offline APK signatures](format-inspectors.md#offline-apk-signatures)
-- <a id="packages-and-extraction"></a>[Packages and extraction](packages-and-archives.md#packages-and-extraction)
-- <a id="offline-pe-resource-inventory"></a>[Offline PE resource inventory](format-inspectors.md#offline-pe-resource-inventory)
-- <a id="offline-pe-and-elf-export-tables"></a>[Offline PE and ELF export tables](format-inspectors.md#offline-pe-and-elf-export-tables)
-- <a id="godot-pck"></a>[Godot PCK](packages-and-archives.md#godot-pck)
-- <a id="esp-images"></a>[ESP images](firmware-and-raw-images.md#esp-images)
-- <a id="esp-flash-images"></a>[ESP flash images](firmware-and-raw-images.md#esp-flash-images)
-- <a id="bun-standalone-executables"></a>[Bun standalone executables](packages-and-archives.md#bun-standalone-executables)
-- <a id="release-exposure"></a>[Release exposure](format-inspectors.md#release-exposure)
-- <a id="windows-shell-link-shortcuts"></a>[Windows Shell Link shortcuts](format-inspectors.md#windows-shell-link-shortcuts)
-- <a id="elf-and-mach-o-initializer-functions"></a>[ELF and Mach-O initializer functions](format-inspectors.md#elf-and-mach-o-initializer-functions)
-- <a id="elf-unwind-function-ranges"></a>[ELF unwind function ranges](format-inspectors.md#elf-unwind-function-ranges)
-- <a id="linux-kernel-build-config"></a>[Linux kernel build config](format-inspectors.md#linux-kernel-build-config)
-- <a id="deno-compile-executables"></a>[Deno compile executables](packages-and-archives.md#deno-compile-executables)
+## Offline PE and ELF export tables
+
+Use `inspect_native_exports` with an absolute `path`, or
+`rea inspect-native-exports ./libexample.so --json`, to read a PE export
+directory or ELF dynamic symbol table on any host without opening a native
+target. Ordinal-only, forwarded and hidden-version exports stay distinct;
+partial or unsupported coverage means the inventory is incomplete, not that
+exports are absent. Use `inspect_macho` for Mach-O. See
+[offline binary diagnostics](https://github.com/morluto/rea/blob/main/docs/guides/native/binary-diagnostics.md).
+
+For exact core-module import/export differences, use `compare_wasm_interfaces`
+with two complete inspection Evidence records (or current-session retained
+references) and explicit scope. Preserve unknown types/duplicate names; static
+interface equality is not runtime compatibility. CLI accepts complete inline
+Evidence in `rea compare-wasm-interfaces input.json --json`, or exact references
+with `--evidence-bundle /absolute/path/evidence.json`. Use the complete parent
+ID when the inspection returned a summary.
+
+## Godot PCK
+
+For a standalone PCK, `open_binary` then `inspect_artifact` with
+`detail: "summary"` retains the complete static inventory. Use the returned
+`parent_evidence_id` with `inspect_analysis_view` inventory occurrences filtered
+by exact `logical_paths` or `occurrence_ids`, and join `artifact_id` to nodes
+filtered by `artifact_ids`. Select only needed fields; identity links remain.
+Prefer `inspect-artifact --detail summary --evidence_output PATH` for CLI tasks;
+the same CLI discovers embedded packs via their EOF trailer without opening a
+native analysis provider. Normal native opening remains unchanged.
+
+Distinguish declared MD5 from observed SHA-256 and graph/page completeness from
+`source_coverage`. Encrypted directories and sparse bundles have unknown entry
+coverage; encrypted members have unknown content identity. An empty filtered
+page cannot prove absence in unknown coverage. Retained views are historical,
+not fresh filesystem checks. No decryption, Godot resource decoding, GDScript
+decompilation or target execution is provided. See
+[Godot PCK](https://github.com/morluto/rea/blob/main/docs/godot-pck.md).
+
+UnityFS AssetBundle containers (`.bundle`, `.ab`, YooAsset packages and other
+byte-identified files) list `CAB-*` SerializedFiles and `.resS` members with
+header and node facts; none/LZ4/LZ4HC member blocks stream on demand. Bundles
+declaring asset-bundle encryption keep header facts with unknown directory
+coverage; LZMA member bytes retain named occurrences with unavailable hashes.
+Virtual member offsets are not physical carrier locations. Format 6–8 readers
+cache one decoded block and support retained filtered/paged views. No
+decryption or SerializedFile object decoding is provided. See
+[UnityFS](https://github.com/morluto/rea/blob/main/docs/unityfs.md).
+
+WeChat mini-program packages (`.wxapkg` files with `0xBE`/`0xED` marks) list
+members with declared names and absolute carrier offsets; member bytes are
+stored verbatim. `V1MMWX` PC-client packages report `encrypted` directory
+status with unknown coverage and are refused on extraction; no decryption is
+attempted. See
+[wxapkg](https://github.com/morluto/rea/blob/main/docs/wxapkg.md).
+
+U-Boot legacy uImages use the same inspection/extraction workflows. Prefer a
+retained summary and selected inventory views for header facts or component
+IDs. Header addresses/CPU/OS are declarations; payloads remain as stored.
+Multi-file components retain their aligned physical ranges; the full payload
+is also materialized. Unknown name text retains its original 32 bytes. See
+[U-Boot uImage](https://github.com/morluto/rea/blob/main/docs/uimage.md).
+
+## ESP images
+
+For an ESP32-family app or bootloader image, `open_binary` then
+`inspect_artifact` with `detail: "summary"`. `root_format_metadata` holds the
+observed `header` and `descriptor` (its `kind` is the app or bootloader role)
+and the derived `chip`, `checksum`, `appended_sha256` and
+`undescribed_trailing` ranges. Segment rows (`segment-NN.bin`) carry
+`load_addr`, `classification` (`mapped`, `loaded`, `not-loaded` or `unknown`,
+per ESP-IDF `should_map`/`should_load`; RAM ranges are not checked) and
+`source_location`; select them with `inspect_analysis_view` and
+`logical_paths`. Signature blocks are `unverified`; `public_key_sha256`
+matches eFuse key digests.
+
+Unknown `source_coverage` means truncation or undescribed trailing bytes, such
+as the flash segments of a `--ram-only-header` image; do not infer absence
+from it. Erased flash is described. Integrity mismatches are facts, not
+refusals. ESP8266 images are refused, and flash-encrypted images are not
+recognized. Segments are not disassembled, and RISC-V or Xtensa segments
+cannot be opened as raw-binary. See
+[ESP images](https://github.com/morluto/rea/blob/main/docs/esp-image.md).
+
+## ESP flash images
+
+A whole-flash dump opens as `esp-flash-image`. `root_format_metadata` holds
+`partition_table` (its `method`: `default`, `caller` or a derived
+`scan`), `bootloader` and `apps` ESP image facts, the `otadata` records,
+`partitions_outside_dump` and `boot_selection`. Every table has a
+`boot_selection`: its `basis` names the bootloader's branch (`no-otadata`
+without otadata, `none` when the bootloader rejects the table, `unknown`
+when a record is missing), and it lists its `assumptions`;
+`running_partition` is always `unknown`. Each image names its entry
+`path`. With `detail: "summary"` the root kind is `esp-flash-image-summary`
+and each image is a compact `summary` (chip, entry point, segment count,
+classified load ranges, signature block count, descriptor identity,
+integrity); complete detail keeps every segment. Partition rows
+(`partitions/<label>.bin`) carry the declared `type`, `subtype`, names,
+`flash_address`, `size` and flags; `encrypted` is a storage flag, not
+proof of ciphertext. `written_offset`/`written_end` bound the bytes that
+are not erased: an app slot whose facts say `erased` can still hold an older
+image, and a limitation says so. For a dump read from another flash address, call
+`open_binary` with `format: "esp-flash-image"` and `flash_offset`; an
+`esp-image` whose `flash_dump` is set names that offset in a next probe; a
+relocated table can be named with `partition_table_offset`. No table found
+means unknown, often flash encryption. Extraction writes `nvs_keys` as
+ordinary data. See
+[ESP flash images](https://github.com/morluto/rea/blob/main/docs/esp-flash-image.md).
+
+## Bun standalone executables
+
+A `bun build --compile` executable stores its module graph in Mach-O
+`__BUN,__bun`, a PE or ELF `.bun` section, or an appended trailer.
+`open_binary` then `inspect_artifact` reports root format `bun-standalone` with
+the carrier, structural layout, module count, entry point, raw flags and
+compile exec-argv; each module keeps its `/$bunfs/...` source name, loader,
+encoding, sizes and SHA-256, with sourcemap, bytecode and module-info regions as
+separate children. Use summary root metadata `entry_point_logical_path` as the
+exact inventory `logical_paths` selector, then read the selected content node by
+its artifact ID; the reader owns virtual-name normalization and indexed fallback. `extract-artifact` writes stored contents through the safe
+output tree; route extracted JavaScript to `analyze_javascript_application`.
+Layout labels describe record shapes, not a detected compiler version; unknown
+or ambiguous layouts are unsupported. Bytecode is retained, not decoded, and the
+executable is never run. See
+[Bun standalone executables](https://github.com/morluto/rea/blob/main/docs/bun-standalone.md).
+
+## Deno compile executables
+
+A `deno compile` executable stores its application in a Mach-O `__SUI,d3n0l4nd`
+section, an Intel Mach-O `<~sui-data~>` trailer, an ELF `SUI` note of type
+`0x53554901`, or a PE `RT_RCDATA` resource named `D3N0L4ND`.
+`open_binary` then `inspect_artifact` reports root format `deno-standalone`
+with the carrier, entry point, recorded options, and one occurrence per virtual
+file, including size and SHA-256. Transpiled output, CJS export analysis, and
+source maps are separate occurrences with the file's `index` and their own
+`role`. They normally use a `.deno-*` suffix on the file path and move to an
+indexed path on a collision, so use the reported `logical_path`. Symlink targets
+stay recorded path parts. A virtual filesystem that cannot be listed leaves
+coverage unknown.
+Environment-file values and CA certificate bytes are retained in local Evidence
+with their counts and byte lengths. The npm snapshot length is reported and not
+decoded. `extract-artifact` writes
+stored files through the safe output tree; route extracted JavaScript to
+`analyze_javascript_application`. Other layouts are unsupported, and the
+executable is never run. See
+[Deno compile executables](https://github.com/morluto/rea/blob/main/docs/deno-standalone.md).

.agents/skills/reverse-engineer-anything/references/packages-and-archives.md

@@ -1,146 +0,0 @@
-# Packages and archives
-
-Read only the sections needed for the selected task.
-
-- [Packages and extraction](#packages-and-extraction)
-- [Godot PCK](#godot-pck)
-- [Bun standalone executables](#bun-standalone-executables)
-- [Deno compile executables](#deno-compile-executables)
-
-## Packages and extraction
-
-For PyInstaller executables, the CArchive is the application layer; the native
-bootloader is only its prefix. Prefer summary delivery for large inventories,
-then retained inventory views with selected member fields/pages. Runtime options
-are metadata, not files. Compact delivery still inventories and hashes members.
-A small complete inventory can cost less than extra calls and schemas. See the
-[Python packaged applications guide](https://github.com/morluto/rea/blob/main/docs/python-packaged-applications.md).
-
-Use `inspect_artifact` for application bundles, archives, ZIP/APK/IPA/MSIX/AppX,
-ASAR, or DMG inputs when the artifact graph and findings help answer the
-question. It returns the complete artifact graph inline. Cite graph manifest
-IDs when using them.
-
-For an IPA or a macOS `.app`, ZIP, or DMG, pass the inventory Evidence to
-`project_apple_application_graph`. It reports application roots and nested
-bundles by path convention: app extensions, XPC services, frameworks, login
-items, system and driver extensions, and plug-ins. It also lists privileged
-helpers, launchd plists, symlinks, and each bundle's `info_plist_path` and
-executable candidates. Roles are path conventions, not parsed plists; read the
-listed plists with `inspect_plist`.
-
-After `open_binary` on a `.app` or Mach-O, use `trace_dylib_resolution` to see
-which file each `@rpath`, `@loader_path` and `@executable_path` load reaches for
-every executable in the bundle. It also lists missing or weak loads and earlier
-`@rpath` candidates that are absent. Narrow large bundles with `roots` or
-`architecture`. System paths stay undetermined because the dyld shared cache
-provides them.
-
-Use `extract_artifact` when materialized files from a supported format are needed.
-It extracts the active target and accepts optional `integrity_policy`; see the [MCP contracts](https://github.com/morluto/rea/blob/main/docs/reference/mcp-contracts.md).
-It materializes all regular files into a fresh temporary directory chosen by
-REA. Symlinks and encrypted entries are inventory facts, not extractable files.
-The result reports the materialized directory; callers do not choose the
-destination. No separate permission grant is needed.
-
-For classic iNES cartridges, `open_binary` recognizes the byte signature without
-an executable architecture. `inspect_artifact` retains header facts and each
-PRG/CHR file unit's offset, length and SHA-256; its nested inventory's
-`ines.reconstruction` verifies the actual ordered region bytes against the
-whole-file digest. Extraction includes the original header, optional trainer
-and uninterpreted trailing bytes. These file units are not mapper/CPU bank
-maps. NES 2.0, UNIF, dirty legacy headers and truncated payloads are explicitly
-refused. No emulator or 6502 analysis is performed. See the
-[iNES guide](https://github.com/morluto/rea/blob/main/docs/ines-containers.md).
-
-Native DMG traversal automatically uses a read-only mount on supported macOS
-hosts, with an owned temporary mount directory and cleanup. It needs no approval
-flag. Unsupported hosts or mount failures remain explicit limitations; do not
-claim child inventory when only root identity is available. DMG extraction is
-unavailable through `extract_artifact`. To access DMG files on macOS, create a
-read-only mount with your own tools, select files from that mount for analysis,
-and detach the mount you created when finished. REA's inspection mount is
-temporary, not a retained export of the image's contents.
-
-Use `inspect_byte_entropy` to triage high- and low-entropy ranges before extraction.
-
-## Godot PCK
-
-For a standalone PCK, `open_binary` then `inspect_artifact` with
-`detail: "summary"` retains the complete static inventory. Use the returned
-`parent_evidence_id` with `inspect_analysis_view` inventory occurrences filtered
-by exact `logical_paths` or `occurrence_ids`, and join `artifact_id` to nodes
-filtered by `artifact_ids`. Select only needed fields; identity links remain.
-Prefer `inspect-artifact --detail summary --evidence_output PATH` for CLI tasks;
-the same CLI discovers embedded packs via their EOF trailer without opening a
-native analysis provider. Normal native opening remains unchanged.
-
-Distinguish declared MD5 from observed SHA-256 and graph/page completeness from
-`source_coverage`. Encrypted directories and sparse bundles have unknown entry
-coverage; encrypted members have unknown content identity. An empty filtered
-page cannot prove absence in unknown coverage. Retained views are historical,
-not fresh filesystem checks. No decryption, Godot resource decoding, GDScript
-decompilation or target execution is provided. See
-[Godot PCK](https://github.com/morluto/rea/blob/main/docs/godot-pck.md).
-
-UnityFS AssetBundle containers (`.bundle`, `.ab`, YooAsset packages and other
-byte-identified files) list `CAB-*` SerializedFiles and `.resS` members with
-header and node facts; none/LZ4/LZ4HC member blocks stream on demand. Bundles
-declaring asset-bundle encryption keep header facts with unknown directory
-coverage; LZMA member bytes retain named occurrences with unavailable hashes.
-Virtual member offsets are not physical carrier locations. Format 6–8 readers
-cache one decoded block and support retained filtered/paged views. No
-decryption or SerializedFile object decoding is provided. See
-[UnityFS](https://github.com/morluto/rea/blob/main/docs/unityfs.md).
-
-WeChat mini-program packages (`.wxapkg` files with `0xBE`/`0xED` marks) list
-members with declared names and absolute carrier offsets; member bytes are
-stored verbatim. `V1MMWX` PC-client packages report `encrypted` directory
-status with unknown coverage and are refused on extraction; no decryption is
-attempted. See
-[wxapkg](https://github.com/morluto/rea/blob/main/docs/wxapkg.md).
-
-U-Boot legacy uImages use the same inspection/extraction workflows. Prefer a
-retained summary and selected inventory views for header facts or component
-IDs. Header addresses/CPU/OS are declarations; payloads remain as stored.
-Multi-file components retain their aligned physical ranges; the full payload
-is also materialized. Unknown name text retains its original 32 bytes. See
-[U-Boot uImage](https://github.com/morluto/rea/blob/main/docs/uimage.md).
-
-## Bun standalone executables
-
-A `bun build --compile` executable stores its module graph in Mach-O
-`__BUN,__bun`, a PE or ELF `.bun` section, or an appended trailer.
-`open_binary` then `inspect_artifact` reports root format `bun-standalone` with
-the carrier, structural layout, module count, entry point, raw flags and
-compile exec-argv; each module keeps its `/$bunfs/...` source name, loader,
-encoding, sizes and SHA-256, with sourcemap, bytecode and module-info regions as
-separate children. Use summary root metadata `entry_point_logical_path` as the
-exact inventory `logical_paths` selector, then read the selected content node by
-its artifact ID; the reader owns virtual-name normalization and indexed fallback. `extract-artifact` writes stored contents through the safe
-output tree; route extracted JavaScript to `analyze_javascript_application`.
-Layout labels describe record shapes, not a detected compiler version; unknown
-or ambiguous layouts are unsupported. Bytecode is retained, not decoded, and the
-executable is never run. See
-[Bun standalone executables](https://github.com/morluto/rea/blob/main/docs/bun-standalone.md).
-
-## Deno compile executables
-
-A `deno compile` executable stores its application in a Mach-O `__SUI,d3n0l4nd`
-section, an Intel Mach-O `<~sui-data~>` trailer, an ELF `SUI` note of type
-`0x53554901`, or a PE `RT_RCDATA` resource named `D3N0L4ND`.
-`open_binary` then `inspect_artifact` reports root format `deno-standalone`
-with the carrier, entry point, recorded options, and one occurrence per virtual
-file, including size and SHA-256. Transpiled output, CJS export analysis, and
-source maps are separate occurrences with the file's `index` and their own
-`role`. They normally use a `.deno-*` suffix on the file path and move to an
-indexed path on a collision, so use the reported `logical_path`. Symlink targets
-stay recorded path parts. A virtual filesystem that cannot be listed leaves
-coverage unknown.
-Environment-file values and CA certificate bytes are retained in local Evidence
-with their counts and byte lengths. The npm snapshot length is reported and not
-decoded. `extract-artifact` writes
-stored files through the safe output tree; route extracted JavaScript to
-`analyze_javascript_application`. Other layouts are unsupported, and the
-executable is never run. See
-[Deno compile executables](https://github.com/morluto/rea/blob/main/docs/deno-standalone.md).

.agents/skills/reverse-engineer-anything/references/recorded-captures.md

@@ -1,58 +0,0 @@
-# Recorded captures and decoder discovery
-
-## Historical web captures
-
-Retained HAR/SAZ/native mitmproxy/aioquic qlog capture: `inspect_web_network_capture`. Select
-original record ordinals when useful; inspect producer fields and byte/number
-sidecars without fetching recorded URLs or claiming live attribution. For a decoder heap refusal, follow the reported `capture.decoder_old_generation_mib`
-setting and bounds with `update_analysis_settings` (default 192 MiB, floor 8,
-maximum 16384); the launch seed
-`REA_CAPTURE_DECODER_OLD_GENERATION_MIB` affects future REA processes. Host memory
-and `NODE_OPTIONS` do not raise that decoder allowance. See
-[historical captures](https://github.com/morluto/rea/blob/main/docs/guides/web/web-network-captures.md) for the exact upstream
-profile and credential exclusions. Qlog support starts with aioquic 1.5.0
-JSON 0.3; stream coverage and endpoint loss declarations do not recover bodies
-or establish network loss. Prefer detail summary, then kind qlog trace/stream
-pages; select one trace, stream ID and direction for contribution/range pages.
-Facts retain all-recorded-events-before-selection scope. Other qlog layouts
-and PCAP correlation are unsupported.
-
-## Packet fields and packet captures
-
-Packet-field discovery: `inspect_packet_fields` searches installed protocols
-and field names/types before choosing fields for `inspect_packet_capture`.
-Query/page the relevant records; registrations do not establish observed traffic.
-
-## Packet capture inspection
-
-Recorded packet captures: `inspect_packet_capture` with selected TShark fields,
-display filters and decode-as rules. Requires Linux, util-linux prlimit and
-TShark 4.x on an absolute `PATH` entry or via the optional absolute
-`REA_TSHARK_COMMAND` override; see
-[capture guide](https://github.com/morluto/rea/blob/main/docs/guides/data/packet-captures.md).
-Packet limits count input packets before filtering. Output defaults to 50 matching
-frames; use `match_offset` / `match_limit` for later pages, or explicit null
-limit for complete matching output. Preserve the source digest across pages.
-This does not perform live capture.
-For address/DNS/SNI/protocol summaries use `inspect_pcap_capture` with MCP
-detail summary, then retained kind pcap pages. Requires Wireshark 4.x tshark
-and capinfos, with Linux prlimit or Windows Job Objects; see
-[packet summaries](https://github.com/morluto/rea/blob/main/docs/guides/data/pcap-analysis.md).
-Select an exact DNS question for address pages; page exhaustion does not prove
-network completeness. This reads recordings and does not acquire traffic.
-
-## Signal decoder discovery
-
-Decoder discovery: `inspect_signal_decoders` searches installed decoder IDs
-or selected decoder help, including channels and options. Query and page the
-relevant details before choosing settings; this does not scan hardware.
-
-## Recorded logic and signals
-
-Recorded logic/signal traces: `decode_signal_recording` with explicit decoder,
-channel and protocol settings. On Linux with util-linux prlimit, supply sigrok-cli 0.7.2 through
-`REA_SIGROK_COMMAND`; see [recording guide](https://github.com/morluto/rea/blob/main/docs/guides/signals/signal-recordings.md).
-Output defaults to 50 literal lines; select annotation classes and page with
-`line_offset` / `line_limit`, or explicit null for all remaining output.
-Window counts are display lines, not annotations or samples. Preserve source digests across pages.
-This reads existing files; it does not acquire hardware signals.

.agents/skills/reverse-engineer-anything/references/runtime-observation.md

@@ -1,12 +1,5 @@
 # Runtime observation
 
-Read only the sections needed for the selected task.
-
-- [Native runtime questions](#native-runtime-questions)
-- [Browser and Electron](#browser-and-electron)
-- [Who performs the action](#who-performs-the-action)
-- [Fiddler Everywhere traffic debugging](#fiddler-everywhere-traffic-debugging)
-
 ## Native runtime questions
 
 Start from the requested observation. State the selected target, input, expected
@@ -61,18 +54,6 @@ and [native calls/UI](https://github.com/morluto/rea/blob/main/docs/guides/nativ
 
 ## Browser and Electron
 
-Passive traffic capture of a SPA: the default scope is the page's own origin,
-and data APIs usually live on another origin. Discover API origins from
-absolute URLs in loaded bundles or script text via JavaScript analysis tools
-(another passive capture without those origins still scopes to the page
-origin, and `completeness.excluded` counts lack URLs). Pass both the page
-origin and those API origins in `allowed_origins` — the list replaces the
-default scope — and read `completeness.excluded` before reporting that an
-API is absent. For long captures with body shapes, use
-`export_evidence_bundle` instead of expecting an inline result.
-
-User-owned Electron runtime already open: `list_electron_targets`. Before an electron-main Inspector attempt, check packaged fuses with `inspect_electron_fuses`; disabled `EnableNodeCliInspectArguments` means the app ignores `--inspect`.
-
 Supply the existing browser or Electron instance's literal-port loopback CDP
 endpoint in the request. No separate permission configuration or per-call
 approval flag is needed. Browser discovery lists eligible HTTP(S) pages;

.cursor-plugin/skills/reverse-engineer-anything/SKILL.md

@@ -2,123 +2,607 @@
 name: reverse-engineer-anything
 description: Reverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.
 metadata:
-  version: "64"
+  version: "62"
 ---
 
 # REA
 
-Use REA when a claim depends on a shipped artifact, decompilation, requested
-runtime evidence, or behavior not established by available source. For ordinary
-analysis of a complete source repository, use normal repository tools and do not run REA
+Use REA when a claim depends on a shipped binary or package, decompilation,
+requested runtime evidence, or comparison with
+behavior not established by available source. For ordinary analysis of a
+complete source repository, use normal repository tools and do not run REA
 readiness or provider commands.
 
-## Select the target and authority
-
-Use the target selected for this task. If the user says only "my app", ask which
-app before searching artifacts. Resolve a named app to one clear installed
-artifact; ask when ambiguous. Unrelated reports and nearby fixtures do not select
-it. Never substitute an example or broaden the selected target or runtime actions.
-
-Distinguish observations, derivations, inferences, and unknowns. Cite Evidence
-IDs and preserve source identity, coverage, limitations and partial facts.
-Static code, strings and declarations do not establish observed execution.
-Within the requested investigation, call available analysis tools directly;
-REA requires no permission grants or per-call approval flags. The client may
-independently require approval; preserve its actual access requirements.
-
 ## Connect only when needed
 
-If REA tools are available and registration is not known stale, proceed directly.
-Use the connected server's actual tool list and schemas: repository-main guidance
-may describe features absent from an older release. Keep complete inline Evidence
-when retained references are not advertised. Skill installation alone adds no
-MCP server or engines; a Claude Code, Codex or Cursor Marketplace plugin bundles its own server.
+Installing this skill supplies instructions; it does not register the REA MCP
+server, install analysis engines, or add tools to an already-running agent.
+The exception is a Claude Code, Codex, or Cursor Marketplace plugin install:
+the plugin bundles this skill and registers its own `rea` server, outside the
+state that `rea setup` and `rea doctor` manage.
 
-For absent tools, blocked calls, unavailable providers, host timeouts or startup
-failures, read [connection and recovery](references/connection-and-recovery.md)
-before repair. Diagnose the observed blocker, continue useful supported work,
-and resume the original task after repair. Do not repeat unchanged failed calls.
-A host timeout without a REA result does not establish cancellation or cleanup.
-Retry a dropped connection once only if safe to repeat or known not started;
-if it may have changed the target, or fails again, ask for a client restart.
-Retry failed connection startup once; do not retry the same unchanged failure.
+If REA tools are available and their registration is not known to be stale,
+proceed directly to the target. Do not run diagnostics or setup before every
+investigation. Use the connected server's actual tool list and input schemas;
+a skill installed from repository main may describe capabilities absent from
+an older npm release. Keep complete inline Evidence when that server does not
+advertise retained references.
 
-Choose the registration owner: plugin recovery differs from setup-owned repair.
+If tools are absent, a call is blocked by the client, or a required provider is
+unavailable, read [connection and recovery](references/connection-and-recovery.md).
+Diagnose the observed blocker, continue useful supported work, and resume the
+original task after repair. Client approval, shell PATH, registration and engine
+availability have different remedies; avoid repeating setup for all of them.
+If the host times out without a REA result, do not repeat the unchanged call or
+assume work stopped. Read the recovery guide's deadline distinction before retrying.
+If the MCP connection drops, retry that call once only when it is safe to
+repeat or known not to have started. If it may already have changed the
+target, or it fails again, tell the user and ask them to restart the MCP
+client. If REA could not start its connection, retry once, then tell the user
+and ask them to restart the MCP client. Do not retry a startup failure that
+returns the same unchanged result. For plugin installs, follow
+[plugin recovery](references/connection-and-recovery.md#rea-installed-as-a-client-plugin).
+For setup-owned registrations, run `rea doctor` when REA could not start
+or tools stay missing after that restart. For `cleanup_incomplete`, a retryable
+capacity limit, or `configuration_invalid`, follow the recovery guide's ladder
+before asking for a restart. Read that guide before a second repair attempt.
+
+Choose the installation owner before repairing registration. Plugin installs use
+[plugin recovery](references/connection-and-recovery.md#rea-installed-as-a-client-plugin);
+only setup-owned registrations use the guide's
+[registration repair plan](references/connection-and-recovery.md#repair-registration-when-needed).
 A plugin-only doctor result of `missing` does not call for `rea setup`.
-Setup-owned repair starts with a read-only plan showing paths, backups and skill
-replacement before authorized writes. Do not install unrelated software to repair
-registration. REA setup never installs/upgrades Node, npm, Homebrew, Java,
-Ghidra, IDA, JADX, Kuna, Binwalk or Unblob.
-For an unsupported client, use manual stdio with an existing absolute runtime
-and entry script, or the CLI fallback in the recovery guide.
 
-## Route before analysis
+REA setup never installs or upgrades Node.js, npm, Homebrew, Java, Ghidra, IDA,
+JADX, Kuna, Binwalk, or Unblob. Use existing prerequisites; do not install unrelated
+software to repair MCP registration. For an unsupported client, use manual stdio
+registration with an existing installation’s absolute runtime and entry script, or continue through the CLI.
 
-Read the matching guide before its workflow; read only the relevant sections.
-Target-free tools take their explicit paths/endpoints without `open_binary`.
-Active-target native/archive tools require it. Archive members and executable
-code are different questions; follow the selected artifact's application layer.
+A concrete CLI fallback for an operator-supplied JavaScript tree or ASAR is:
 
-| Supplied target or question                                                                                                                                        | Guide and first operation                                                                                                                                                                                                                          |
-| ------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| Native executable/library, Ghidra/IDA/Kuna, managed PE/CLI or NativeAOT                                                                                            | [Native and managed](references/native-and-artifacts.md): `open_binary`; managed CLI metadata: `inspect_managed_artifact`. Read provider/lifecycle sections before reuse or edits.                                                                 |
-| Headerless/raw firmware, Cortex-M, ESP app/bootloader or whole-flash dump, HEX/S-record/UF2/DfuSe                                                                  | [Firmware and raw images](references/firmware-and-raw-images.md): check loader support first; raw: explicit `open_binary` format/mapping; ESP/container: `open_binary` then `inspect_artifact`; region scan/extraction: advertised firmware tools. |
-| Archive/package inventory or extraction, PyInstaller, Bun/Deno executable, Godot PCK, UnityFS or wxapkg                                                            | [Packages and archives](references/packages-and-archives.md): `open_binary`, then `inspect_artifact`; read extraction/mount limits before materializing.                                                                                           |
-| ASAR/extracted JavaScript or Electron tree, feature trace or version comparison                                                                                    | [JavaScript applications](references/javascript-applications.md): `analyze_javascript_application`.                                                                                                                                                |
-| APK code, manifest, classes, methods or incoming references                                                                                                        | [Android applications](references/android-applications.md): `inspect_android_package`; signing/inventory use their separate routes.                                                                                                                |
-| PE/ELF/Mach-O tables or hardening, Go/Rust/IL2CPP/GPU/model metadata, Windows shortcuts, kernel config, pyc/BEAM, SQLite, WASM, EVM, APK signing or recorded crash | [Offline format inspectors](references/format-inspectors.md): select the named target-free inspector in that format's section.                                                                                                                     |
-| Shipped application exposure for a local app, install directory, ASAR or executable                                                                                | [Offline format inspectors](references/format-inspectors.md#release-exposure): `audit_release_exposure`; preserve unknown/unassessed checks and marker-only credential findings.                                                                   |
-| Apple plist, Interface Builder, asset catalog or keyed archive resources; unknown desktop installation directory                                                   | [Resource and directory inspection](references/native-and-artifacts.md#apple-app-resources): `open_bundle_resources` or `inspect_desktop_application`; no executable/provider required for resource sessions.                                      |
-| Native execution, live calls, instrumentation, process effects; browser/Electron runtime or agent-performed actions                                                | [Runtime observation](references/runtime-observation.md): select the observation and declared lifecycle first. Already-open page: `list_browser_targets`; Electron: `list_electron_targets`; agent actions: `capture_browser_scenario`.            |
-| Recorded HAR/SAZ/mitmproxy/qlog, packet capture, protocol-field discovery or logic/signal recording                                                                | [Recorded captures](references/recorded-captures.md): `inspect_web_network_capture`, packet/signal inspection or decoder discovery as appropriate; no live acquisition or recorded-URL fetch.                                                      |
-| Web AgentEnv bundle from a live instance or capture; task for an exported desktop environment                                                                      | [Environments](references/environments.md), including its desktop-task section: read before recording/export; `capture_browser_scenario` then `export_environment_bundle` for a live web bundle.                                                   |
-| Linux desktop application environment built from a user request                                                                                                    | [Desktop builds](references/desktop-build.md): drive the selected app on a build computer, then export and witness.                                                                                                                                |
-| Retained Evidence paging/composition, comparisons, unknowns or a broader investigation                                                                             | [Evidence workflows](references/evidence-workflows.md): use existing source IDs and the smallest view that answers the question.                                                                                                                   |
+```bash
+npx -y rea-agents@latest analyze-javascript-application /absolute/path/to/app --json
+```
 
-If a needed tool is unadvertised, inspect `binary_session {}` and
-`result.tool_availability` for availability reasons/remediation; never call an
-unadvertised tool. Expand group `names`, inheriting inventory client requirements
-unless overridden. Their union is the complete in-scope inventory; `tools/list`
-owns canonical schemas. The default summary keeps provider candidate status,
-code and reason; `omitted_fields` names exclusions. Request `detail: "complete"`
-only for remediation/diagnostics, candidate capabilities or tool effect flags.
-Candidate facts describe possible providers; top-level capabilities and
-`analysis_provider_binding` describe the current route. Keep those scopes distinct.
+No MCP registration or native engine is required. Read the returned Evidence,
+graph, limitations, and unknowns with the same care as an MCP result; the CLI
+returns the Evidence record directly. This fallback does not establish that
+MCP is configured. Native CLI tasks still require their selected engine.
+Prefer an existing `rea` CLI. When none is installed and REA package acquisition
+is within the authorized scope, use `npx -y rea-agents@latest` in its place from a
+trusted directory outside the target project, with absolute target paths. Managed
+setup binds the current installation directly; retain its runtime and package
+cache entry, or restore that installation before repairing its registration.
+For setup-owned registration, configuration repair starts with a read-only setup plan; show paths, backups and
+skill replacement before applying authorized changes. Preserve existing approval
+and target-execution boundaries.
+
+## Route the target first
+
+Choose the first tool from the target the user supplied. Use `open_binary` for
+active-target native or archive workflows; target-free tools take their own
+explicit path or endpoint and do not need it.
+
+- ASAR or extracted JavaScript/Electron tree:
+  `analyze_javascript_application`.
+- Archive/package member inventory (ZIP/APK/IPA/MSIX/AppX, DMG, or a PyInstaller
+  executable): `open_binary` with the supplied local path; use `inspect_artifact`
+  when its graph and findings help answer the question. A PyInstaller CArchive
+  is the application layer; the bootloader is only the native prefix. Start with
+  `detail: "summary"` for large inventories, then use retained inventory views
+  for selected member fields/pages. A complete small inventory can cost less
+  than extra calls and schemas. Runtime options are metadata, not files.
+  Compact delivery still inventories and hashes members. See the
+  [Python packaged applications guide](https://github.com/morluto/rea/blob/main/docs/python-packaged-applications.md).
+- Bun `--compile` executable: `open_binary`, then `inspect_artifact`; the Bun
+  module graph is the application layer. Extract stored modules and analyze the
+  JavaScript. See [Bun standalone executables](https://github.com/morluto/rea/blob/main/docs/bun-standalone.md).
+- Deno `deno compile` executable: `open_binary`, then `inspect_artifact`; the
+  embedded virtual filesystem is the application layer. Extract stored files and
+  analyze compatible JavaScript. Environment-file values and CA certificate bytes
+  are retained in local Evidence with their counts and byte lengths. The npm
+  snapshot length is reported and not decoded, and the executable is not run. See
+  [Deno compile executables](https://github.com/morluto/rea/blob/main/docs/deno-standalone.md).
+- Apple app plist, Interface Builder, asset catalog, or keyed archive resources:
+  `open_bundle_resources`, then the selected resource tool. This does not
+  require `CFBundleExecutable` or a deep provider. Resource sessions have no
+  executable digest or binary snapshot; use `open_binary` for native code.
+- Android APK code, classes, methods, or incoming references:
+  `inspect_android_package`, then focused Android tools when advertised.
+  Archive member inventory still uses the archive route above.
+- Managed PE/CLI assembly: `inspect_managed_artifact`.
+- Firmware image: `inspect_firmware_regions` when advertised. Use
+  `extract_firmware` when extraction is requested, with a caller-selected new
+  absolute output directory. These tools use caller-supplied Binwalk/Unblob on
+  Linux; see the [firmware guide](https://github.com/morluto/rea/blob/main/docs/guides/firmware/firmware-analysis.md).
+- ESP32-family app or bootloader image (`0xE9` magic): `inspect_artifact`
+  without a provider; see the ESP section of `references/native-and-artifacts.md`.
+- .NET NativeAOT PE/ELF: native Ghidra analysis. Native code can yield pseudocode;
+  consult the NativeAOT workflow in `references/native-and-artifacts.md` for
+  optional metadata recovery and its supported host/layout boundary.
+- Native program behavior, live calls, instrumentation, or command-run effects:
+  choose the matching [native runtime workflow](references/runtime-observation.md#native-runtime-questions).
+  `open_binary` and `trace_native_values` establish static facts; they do not
+  observe execution. Runtime tools have distinct launch, attach, mutation and
+  cleanup effects, and do not require a preliminary static session unless their
+  own contract says so.
+- User-owned browser page already open: `list_browser_targets`.
+- Passive traffic capture of a SPA: the default scope is the page's own origin,
+  and data APIs usually live on another origin. Discover API origins from
+  absolute URLs in loaded bundles or script text via JavaScript analysis tools
+  (another passive capture without those origins still scopes to the page
+  origin, and `completeness.excluded` counts lack URLs). Pass both the page
+  origin and those API origins in `allowed_origins` — the list replaces the
+  default scope — and read `completeness.excluded` before reporting that an
+  API is absent. For long captures with body shapes, use
+  `export_evidence_bundle` instead of expecting an inline result.
+- Browser action the agent must perform (click, type, navigate):
+  `capture_browser_scenario` with declared actions. See
+  [runtime observation](references/runtime-observation.md) for connect/launch
+  and the focused follow-up browser tools.
+- Agent environment (AgentEnv bundle) of a web application: read
+  [environments](references/environments.md) before recording; it records with
+  `capture_browser_scenario` and writes the bundle with `export_environment_bundle`.
+- Task for a desktop application's exported environment, from a
+  `capture-desktop-application` folder: read
+  [desktop application tasks](references/environments.md#tasks-for-a-desktop-application-environment).
+- Environment of a Linux desktop app with a task, from a request such as
+  "build me an Inkscape environment where the agent changes a poster's
+  headline": read [desktop builds](references/desktop-build.md). You drive the
+  app on a build computer from `start_build_computer`, then export and witness.
+- Fiddler Everywhere traffic or a native `.saz` capture: read the
+  [Fiddler workflow](references/runtime-observation.md#fiddler-everywhere-traffic-debugging).
+  Inspect saved SAZ with `inspect_web_network_capture` and `format: "saz"`;
+  connect Everywhere's separate MCP server for live traffic.
+- Retained HAR/SAZ/native mitmproxy/aioquic qlog capture: `inspect_web_network_capture`. Select
+  original record ordinals when useful; inspect producer fields and byte/number
+  sidecars without fetching recorded URLs or claiming live attribution. HAR/SAZ/qlog
+  old-generation heap is `REA_CAPTURE_DECODER_OLD_GENERATION_MIB` (default 192 MiB,
+  floor 8, maximum 16384), set where REA starts. Host memory and `NODE_OPTIONS` do not
+  raise it. See
+  [historical captures](https://github.com/morluto/rea/blob/main/docs/guides/web/web-network-captures.md) for the exact upstream
+  profile and credential exclusions. Qlog support starts with aioquic 1.5.0
+  JSON 0.3; stream coverage and endpoint loss declarations do not recover bodies
+  or establish network loss. Prefer detail summary, then kind qlog trace/stream
+  pages; select one trace, stream ID and direction for contribution/range pages.
+  Facts retain all-recorded-events-before-selection scope. Other qlog layouts
+  and PCAP correlation are unsupported.
+- User-owned Electron runtime already open: `list_electron_targets`. Before an electron-main Inspector attempt, check packaged fuses with `inspect_electron_fuses`; disabled `EnableNodeCliInspectArguments` means the app ignores `--inspect`.
+- Shipped application exposure for a local `.app`, install directory, ASAR, or executable: `audit_release_exposure` / `rea audit-release-exposure`. Each check has a stable id, a status, and fixed remediation. Unknown and unassessed checks stay in the report. This does not establish that an application is secure. Credential-marker hits report the rule id and location only. See [release exposure](https://github.com/morluto/rea/blob/main/docs/release-exposure.md).
+- Packet-field discovery: `inspect_packet_fields` searches installed protocols
+  and field names/types before choosing fields for `inspect_packet_capture`.
+  Query/page the relevant records; registrations do not establish observed traffic.
+- Recorded packet captures: `inspect_packet_capture` with selected TShark fields,
+  display filters and decode-as rules. Requires Linux, util-linux prlimit and
+  TShark 4.x on an absolute `PATH` entry or via the optional absolute
+  `REA_TSHARK_COMMAND` override; see
+  [capture guide](https://github.com/morluto/rea/blob/main/docs/guides/data/packet-captures.md).
+  Packet limits count input packets before filtering. Output defaults to 50 matching
+  frames; use `match_offset` / `match_limit` for later pages, or explicit null
+  limit for complete matching output. Preserve the source digest across pages.
+  This does not perform live capture.
+  For address/DNS/SNI/protocol summaries use `inspect_pcap_capture` with MCP
+  detail summary, then retained kind pcap pages. Requires Wireshark 4.x tshark
+  and capinfos, with Linux prlimit or Windows Job Objects; see
+  [packet summaries](https://github.com/morluto/rea/blob/main/docs/guides/data/pcap-analysis.md).
+  Select an exact DNS question for address pages; page exhaustion does not prove
+  network completeness. This reads recordings and does not acquire traffic.
+- Decoder discovery: `inspect_signal_decoders` searches installed decoder IDs
+  or selected decoder help, including channels and options. Query and page the
+  relevant details before choosing settings; this does not scan hardware.
+- Recorded logic/signal traces: `decode_signal_recording` with explicit decoder,
+  channel and protocol settings. On Linux with util-linux prlimit, supply sigrok-cli 0.7.2 through
+  `REA_SIGROK_COMMAND`; see [recording guide](https://github.com/morluto/rea/blob/main/docs/guides/signals/signal-recordings.md).
+  Output defaults to 50 literal lines; select annotation classes and page with
+  `line_offset` / `line_limit`, or explicit null for all remaining output.
+  Window counts are display lines, not annotations or samples. Preserve source digests across pages.
+  This reads existing files; it does not acquire hardware signals.
+- Explicit local WebAssembly artifact: `inspect_wasm_artifact`. Supply WABT 1.0.42
+  through `REA_WABT_BIN_DIRECTORY`; retain exact byte/tool identity and decoded WAT.
+  Select JavaScript glue and candidate paths explicitly and preserve URL/path
+  ambiguity. No implicit fetch or target execution; WAT is not original source.
+  See the [WASM guide](https://github.com/morluto/rea/blob/main/docs/guides/wasm/wasm-artifacts.md).
+- Explicit local Android APK (or ZIP/AAB) signing: `inspect_apk_signature`.
+  Prefer its summary or selected signer/digest/lineage page; retain complete
+  Evidence and use `inspect_analysis_view` kind `apk-signature` for follow-ups.
+  Reports which schemes are present (v1 JAR, v2, v3, v3.1), each signer's
+  certificate SHA-256 fingerprint (v1 included, from its PKCS#7 block), v3 key-rotation lineage, source stamp, and
+  whether each signed digest still matches the bytes (self-consistency). This
+  is identity and integrity, not a trust or authenticity verdict.
+  See the [APK signature guide](https://github.com/morluto/rea/blob/main/docs/apk-signature.md).
+- Explicit local EVM bytecode carrier: `inspect_evm_interface` with caller-selected
+  `raw` or `hex` encoding. The owned worker deadline is `timeout_ms` (default
+  30000 ms, range 1..2147483647, next call). Select byte, address-space and CPU
+  allowances separately through `producer_limits`; tighter inherited OS limits
+  remain enforced, and the 256 MiB worker heap is independent. Preserve
+  carrier/decoded digests and treat selectors,
+  argument strings and mutability as inferred candidates; this performs no chain
+  lookup or target execution. For the complete inferred interface without hex or
+  duplicate stdout, discover support and select `detail: "summary"`; complete
+  Evidence stays retained under its original identity. Use the returned
+  `parent_evidence_id` for exact function or `bytecode_hex`/stdout/stderr pages.
+  Candidate coverage does not establish interface completeness. See the
+  [EVM bytecode guide](https://github.com/morluto/rea/blob/main/docs/guides/evm/evm-bytecode.md).
+- Embedded Go compiler, module, build-setting metadata, or static `(ptr, len)`
+  string literals: `inspect_go_binary` with the explicit local file (pass
+  `include_string_literals: true` and optional `max_string_literals` for bounded
+  inferred static string-header recovery, and page retained rows with
+  `inspect_analysis_view` kind `go` collection `string_literals`). This built-in
+  target-free inspection requires no Go installation or disassembler. Preserve
+  raw module bytes and source ranges; absent metadata does not prove a binary is
+  not Go, static string recovery excludes compiler-inlined stack or immediate
+  strings, and unparsed text records remain explicit. See the
+  [Go binary guide](https://github.com/morluto/rea/blob/main/docs/guides/native/go-binaries.md).
+- Embedded CUDA/HIP code objects: `inspect_gpu_code_objects`; inventory targets
+  first, then decode selected payloads and query retained GPU Evidence.
+- Bundled GGUF or safetensors model declarations: `inspect_model_artifact`
+  with the explicit local file. Default header hashing leaves full-file identity
+  unknown; `hash: "full"` additionally reads all weights for SHA-256. MCP returns
+  retained summaries; optional `view` selects one page in the initial call. Use
+  `inspect_analysis_view` kind `model-artifact` for
+  batched metadata keys, tensor names or type totals, selected fields and pages.
+  Preserve header-scoped indices and array preview omissions; do not infer model
+  behavior from declared architecture or tensor types. See the
+  [model file guide](https://github.com/morluto/rea/blob/main/docs/model-artifacts.md).
+- Unity IL2CPP names, member owners and string literals from `global-metadata.dat`:
+  `inspect_il2cpp_metadata` with the explicit local file; MCP defaults to compact
+  counts. Query sections, images, types, methods, fields, properties, events,
+  parameters or string_literals through `inspect_analysis_view`
+  with the returned `parent_evidence_id`, using exact filters, batch indices and
+  selected fields. Parameters accept `method_indices`, avoiding a scan of all
+  parameter pages. Missing collections in historical Evidence remain unknown.
+  Versions 24, 27, 29, 31 and 39; no Unity, IL2CPP
+  binary or disassembler is run. Unrecognized magic (possibly encrypted) and
+  other versions are unsupported. Optional matching `binary_path` enables static native addresses and type references; query `method_addresses` with indices or addresses/states, preserving every shared-body method and both artifact identities. Without a linked binary native counts remain unknown. Byte-pattern fallback is inferred; no runtime behavior or same-build pairing is proved. See the
+  [IL2CPP metadata guide](https://github.com/morluto/rea/blob/main/docs/il2cpp-metadata.md).
+- Embedded cargo-auditable Rust dependency metadata: `inspect_rust_binary`
+  with the explicit local ELF, PE, or thin Mach-O file. This built-in
+  target-free inspection requires no Rust installation or disassembler. Preserve
+  package order, dependency indices, and the section file range. A missing
+  `.dep-v0` section is null metadata, not proof the binary has no dependencies.
+  A rustc commit hash is an inference from a read-only `/rustc/<40-hex>/` path.
+  For selected crates, use `view: {collection: "packages", names: [...],
+  fields: ["name", "version", "source"]}` on the first call; explicitly include
+  `dependencies` for direct index references. Use `detail: "summary"` for counts
+  and ranges, or retained `inspect_analysis_view` Rust package/edge pages for
+  subsequent filters. Keep the complete parent Evidence accessible; CLI compact
+  delivery requires `--evidence_output`.
+  See the
+  [Rust binary guide](https://github.com/morluto/rea/blob/main/docs/rust-binaries.md).
+- PE Authenticode certificate table: `inspect_pe_certificates` with the explicit
+  local PE32/PE32+ file. This built-in target-free inspection requires no
+  Windows SDK. Preserve WIN_CERTIFICATE ranges, types, payload SHA-256, and
+  PKCS#7 signer/X.509 identity fields, including `subject_key_identifier_hex`
+  when the signer uses that form. A missing security directory is observed
+  absence of that table, not proof the image is unsigned in every catalog sense.
+  Signatures are not verified. See the
+  [PE certificate guide](https://github.com/morluto/rea/blob/main/docs/pe-certificates.md).
+- PE import and delay-load tables: `inspect_pe_imports` with the explicit
+  local PE32/PE32+ file. This built-in target-free inspection requires no
+  Windows SDK. Preserve DLL names, named and ordinal functions, ILT/IAT RVAs,
+  delay-load descriptors, and the pefile import hash of the standard import
+  table. A missing import directory is observed absence of that table, not
+  proof the image never imported APIs through another table. Bound snapshots
+  and loaded IAT contents remain unobserved. See the
+  [PE import guide](https://github.com/morluto/rea/blob/main/docs/pe-imports.md).
+- Windows Shell Link shortcuts: `inspect_windows_shortcut` with the explicit
+  local `.lnk` file. This built-in target-free inspection requires no Windows
+  SDK or COM. Preserve header flags and FILETIMEs, IDList items, LinkInfo
+  volume and network names, StringData arguments and icon location, ExtraData
+  signatures, and Tracker droid facts labelled derived. Unknown IDList items
+  and ExtraData signatures stay raw with SHA-256. The target path is not
+  resolved or executed. See the
+  [Windows shortcut guide](https://github.com/morluto/rea/blob/main/docs/guides/native/windows-shortcut.md).
+- PE TLS callbacks and load configuration: `inspect_pe_load_config` with the
+  explicit local PE32/PE32+ file. This built-in target-free inspection requires
+  no Windows SDK. Preserve callback VAs as strings with their RVA, file offset
+  and section, and the size-covered security cookie, SafeSEH, CFG,
+  EH-continuation and XFG fields. Guard-flag mitigation claims are inferred
+  linker declarations, not proof of enforcement. A missing directory is observed
+  absence of that declaration. See the
+  [PE load configuration guide](https://github.com/morluto/rea/blob/main/docs/pe-load-config.md).
+- ELF, PE, and Mach-O hardening: `inspect_binary_hardening` with the explicit
+  local file. This built-in target-free inspection requires no pwntools or other
+  host tool. One image per Mach-O slice. Each property is `enabled`, `disabled`,
+  `partial`, `unknown`, or `not_applicable`, with the header field or import
+  behind it. A missing stack-protector or fortify import is `unknown`, not
+  proof the protection was omitted. Authenticode and Mach-O code signatures
+  report presence, not validity. Hardened runtime stays unknown and names
+  `inspect_signature`. The image is not executed. See the
+  [binary hardening guide](https://github.com/morluto/rea/blob/main/docs/binary-hardening.md).
+- ELF note tables: `inspect_elf_notes` with the explicit local ELF32/ELF64
+  file. This built-in target-free inspection requires no disassembler. Preserve
+  PT_NOTE and SHT_NOTE ranges, owners, types, descriptor SHA-256, and GNU
+  build-id or ABI-tag fields. A missing table is observed absence of those
+  declarations, not proof the image never carried notes. See the
+  [ELF notes guide](https://github.com/morluto/rea/blob/main/docs/elf-notes.md).
+- Code that runs before `main` or at exit in ELF or Mach-O:
+  `inspect_native_initializers` with the explicit local file (and
+  `architecture` for a universal Mach-O). This built-in inspection requires no
+  disassembler. Preserve slot and target addresses, the `resolution` of each
+  entry, and its limitations. An `unresolved` slot is not a zero pointer, and a
+  missing table is not proof that no constructor exists. See the
+  [initializer guide](https://github.com/morluto/rea/blob/main/docs/guides/native/native-initializers.md).
+
+- ELF unwind function ranges: `inspect_elf_unwind_ranges` with the explicit
+  local ELF32/ELF64 file. This built-in target-free inspection requires no
+  disassembler. Preserve FDE start, end, size, containing section, exact
+  symbol name, CIE index, LSDA, and personality. Page with `offset` and
+  `limit`. A missing `.eh_frame` is observed absence of that declaration,
+  not proof the image has no functions. `.debug_frame` is reported present
+  or absent and is not decoded. CFA instructions are counted and not
+  evaluated. See the
+  [ELF unwind ranges guide](https://github.com/morluto/rea/blob/main/docs/guides/native/elf-unwind-ranges.md).
+- Apple `.ips` crash reports: `inspect_apple_ips` with the explicit local
+  file. This built-in target-free inspection requires no Xcode, atos, or
+  pwntools. Preserve first-line metadata, `bug_type` as a string, process and
+  exception facts, lossless integer registers, usedImages UUIDs, and
+  documented classification matches. Only `bug_type` `"309"` yields a crash
+  body; other types, including `"288"` stackshots, return no crash result.
+  Select an optional `join_path` for UUID-and-architecture matching to one
+  local Mach-O or dSYM; preserve slice identity and unjoined reasons. Joined
+  static addresses do not resolve function names or dSYM symbols. See the
+  [Apple IPS guide](https://github.com/morluto/rea/blob/main/docs/apple-ips.md).
+- Linux kernel module metadata: `inspect_kernel_module` with the explicit local
+  ELF32/ELF64 relocatable file. This built-in target-free inspection requires
+  no Linux host or `modinfo`. Preserve `.modinfo` key order and duplicate keys,
+  `__versions` CRC hex, section file ranges, and the appended signature
+  trailer's PKCS#7 SHA-256. A missing section is observed absence when section
+  names are known, not proof the module never carried that metadata.
+  `struct module` is not decoded. The signature is not validated. gzip, xz, and
+  zstd containers are unsupported; decompress into a separate file with an
+  external codec utility, then inspect the decompressed `ET_REL` image. See the
+  [kernel module guide](https://github.com/morluto/rea/blob/main/docs/kernel-modules.md).
+- Linux kernel version and build config (`CONFIG_*` options such as
+  `CONFIG_SECURITY_SELINUX`): `inspect_kernel_build_config` with the explicit
+  local `vmlinux`, raw `Image`, x86 `bzImage`, or `configs.ko`. This built-in
+  target-free inspection requires no Linux host or `extract-ikconfig`. Preserve
+  the banner version, config SHA-256, option count, and value kinds; use `name`
+  for one option. An `IKCFG_ST` string is a config only with a valid gzip member
+  and `IKCFG_ED`. An absent config may mean the kernel was built without
+  `CONFIG_IKCONFIG`. Non-gzip payloads are unsupported; decompress externally,
+  then inspect the result. See the
+  [kernel build config guide](https://github.com/morluto/rea/blob/main/docs/guides/native/kernel-build-config.md).
+- CPython 3.7–3.13 `.pyc` header, nested marshal code objects and bytecode
+  identity: `inspect_python_pyc` with the explicit local file. This built-in
+  target-free inspection requires no Python installation. Preserve magic,
+  PEP 552 hash or mtime/size, nested names, and bytecode SHA-256. Bytecode is
+  not decompiled, and a decoded header does not prove the matching `.py` is
+  present. See the
+  [CPython pyc guide](https://github.com/morluto/rea/blob/main/docs/python-pyc.md).
+- Erlang BEAM IFF container: `inspect_erlang_beam` with the explicit local
+  `.beam` file. This built-in target-free inspection requires no Erlang
+  installation. Preserve chunk names, atom order, export/import/lambda tables,
+  and the Code header. For selected facts, prefer `detail: "summary"` plus a
+  `beam` view (collection, names/indices, fields and paging), retaining the full
+  Evidence. CLI summary requires `--evidence_output`. Missing table counts stay
+  null; decoded empty tables count zero. Bytecode is not executed. See the
+  [Erlang BEAM guide](https://github.com/morluto/rea/blob/main/docs/erlang-beam.md).
+
+- Explicit local SQLite database snapshot: `inspect_sqlite_database`. Inspect
+  with Node.js 24.x >=24.15 or 26+ providing native SQLite limits. Read schema
+  first (compact inventory by default), or select an exact ordinary table
+  and `columns` when rows are needed. Defaults are selected-table schema and
+  20 rows; request `schema_scope: "database"` for the complete stored schema. Preserve both database and existing sibling WAL identity; row samples
+  report truncation and have unspecified order. Use a quiescent snapshot rather
+  than assuming a live database/WAL pair was copied atomically. See the
+  [SQLite guide](https://github.com/morluto/rea/blob/main/docs/guides/data/sqlite-databases.md).
+- Explicit Linux ELF file for offline layout, symbols, relocations or static
+  mitigation evidence: `inspect_binary_layout`. This target-free operation uses
+  caller-supplied pwntools without opening a disassembler database. Preserve its
+  raw locations and inference/coverage limits; see the
+  [offline binary guide](https://github.com/morluto/rea/blob/main/docs/guides/native/binary-diagnostics.md).
+- Supplied Linux x86-64 ELF core: `inspect_recorded_crash`. Read every recorded
+  thread, raw note and source-bound register without opening a live target.
+  Historical PIDs do not select live processes. Request `include_debugger_context`
+  only when core-only mapping candidates help; see the
+  [recorded crash guide](https://github.com/morluto/rea/blob/main/docs/guides/native/recorded-crashes.md).
+- PE resource type/name/language trees and static icon group references:
+  `inspect_pe_resources` with an explicit path. This portable operation requires
+  no active target or engine; opaque payloads retain file ranges and SHA-256.
+- PE export directories and ELF dynamic-symbol exports: `inspect_native_exports`
+  with an explicit path; portable, no active target or engine.
+- Native executable, library, or analysis database: `open_binary`, then
+  use focused analysis tools directly; call `binary_overview` when metadata or
+  inventory context is useful and available from the selected provider.
+
+For an existing IDA MCP configuration, use `open_binary` with the original
+input binary and `provider_id: "ida"`; do not pass an `.idb` or `.i64` database.
+The legacy attached profile binds the already-open GUI input by SHA-256 and
+leaves its database open. The modern headless profile creates and closes an
+owned private database without saving. Installation and registration reuse
+[mrexodia/ida-pro-mcp](https://github.com/mrexodia/ida-pro-mcp); see the
+[IDA provider guide](https://github.com/morluto/rea/blob/main/docs/providers/ida-provider.md).
+For migration to the separate native IDA MCP plugin, follow the guide's
+[legacy attached migration](https://github.com/morluto/rea/blob/main/docs/providers/ida-provider.md#migrate-a-legacy-attached-registration).
+Replace the upstream registration explicitly; modern mrexodia tools are not the
+native profile. Native attachment preserves the database but does not provide
+direct callers; check session availability and the guide's verification limits.
+Proceed directly to `analyze_function`, pseudocode, or function/string searches
+after opening; IDA does not supply `binary_overview`. Consult session availability
+when composing broader workflows. Modern direct callers and unsupported dossier
+facets remain unknown; live IDA results are not replayed from snapshots.
+
+With `REA_KUNA_COMMAND` configured, `provider_id: "kuna"` answers procedure
+lists, lookups, pseudocode, assembly and `analyze_function` from Kuna's cached
+whole-program export. Callers/callees use a separate body-free graph shared by
+the session; `xrefs` returns exact-address incoming sites and filters forwarding
+aliases. These queries and inventory lookups do not trigger a C export. Indirect
+targets remain unknown and can make relationship coverage partial. Reference
+caches expire on close; a fresh CLI process repeats the native reference walk.
+Typed references, CFG, raw-memory tools and annotations remain unsupported. A cold first
+answer can come from a single-function run (its limitations say so); assembly
+waits for the export. For pseudocode and assembly, `raw_result.source` names the
+cached `.c`/`.asm` file and byte span, which you may search directly instead of
+calling per function. See the
+[Kuna provider guide](https://github.com/morluto/rea/blob/main/docs/kuna-provider.md).
+
+If the user says only "my app" without a name, path, or target already selected
+for this task, ask which app before searching for artifacts. Paths in unrelated
+saved reports and nearby test fixtures do not select the user's target. Resolve
+an actual human-readable app name to one clear installed artifact when possible;
+ask when matches are ambiguous. Never choose an example app on the user's behalf.
+
+In a target-free session, use `open_binary` to bind any archive/package or
+native target whose analysis tool operates on the active target. Do not call an
+unadvertised tool; inspect `binary_session` with `{}` and its
+`result.tool_availability` for availability reasons and remediation. Expand each
+group's `names` into entries, inheriting the inventory-level client requirements
+unless the group overrides them. The grouped name union is the complete tool
+inventory; use `tools/list` for canonical tool schemas. The default summary keeps
+each provider candidate's availability and target-support status, code and
+reason; `omitted_fields` lists what it leaves out. Request `{"detail": "complete"}`
+only when you need an unavailable candidate's remediation, provider
+diagnostics, candidate capability sets, or tool surface, effect and annotation
+flags; in those closed flag vocabularies, listed
+flags are true and omitted known flags are false. Provider candidate facts
+describe possible providers, while top-level session capabilities describe the
+current route and `analysis_provider_binding` names the selected provider. Do
+not treat those scopes as interchangeable.
+
+For editable Ghidra analysis that must survive a later agent process, select
+the [owned Program checkpoint workflow](references/native-and-artifacts.md#continue-editable-ghidra-analysis-across-processes)
+before making metadata edits. It is separate from Evidence/Unknown persistence.
+
+For a Ghidra startup timeout on a large native target, follow
+[deadline recovery](references/native-and-artifacts.md#ghidra-startup-deadline-recovery)
+before abandoning deep analysis or repeating the same cold import. The default
+330 seconds can be changed for new startup attempts through
+`update_analysis_settings` using `ghidra.startup_timeout_ms`. Changing the
+server's launch environment instead requires a new REA process.
 
 ## Work summary-first
 
-Start with inline Evidence and graph context. For unknown/large JavaScript or ELF
-layout, select advertised `detail: "summary"`; MCP `analyze_function` already
-defaults to a summary. Small complete results can cost less than extra pages.
-Make a focused follow-up only for a specific unanswered question. Reuse matching
-session state and Evidence instead of repeating analysis; recheck state when a
-lifecycle change makes it useful.
+Start with the default result and use its inline Evidence and graph context.
+For a JavaScript application or ELF layout whose size is unknown or large,
+call `analyze_javascript_application` or `inspect_binary_layout` with
+`"detail": "summary"`, then inspect the retained `parent_evidence_id` with
+`inspect_analysis_view`. Over MCP, `analyze_function` is summary-first without
+a flag: its default reply is the summary, and the complete dossier stays
+retained under `parent_evidence_id` for facet paging. Pass that ID as a
+`retained-evidence` reference to `compare_functions`, or inside
+`match_native_functions` `left_observations`/`right_observations` or
+`build_call_path.functions`; request `"detail": "complete"` only when the
+dossier must travel inline.
+Use the same exact-reference form in other advertised logical Evidence inputs,
+including optional sources, arrays and nested sources across domains. Each
+source can independently be inline or retained; preserve its source identity,
+coverage, partial facts and unknowns. A summary or inspection page has its own
+Evidence ID: use `normalized_result.parent_evidence_id` when the consumer needs
+the complete source. For portable CLI follow-ups, commands advertising
+`--evidence-bundle` resolve references from that selected exported bundle.
+See [Evidence composition](references/evidence-workflows.md) for scope and
+recovery; file selectors are supported only where explicitly advertised.
+Page module identities and select the returned `node_id` before requesting an
+item; a single file path can represent several graph roles. Do not repeat an
+identical analysis or rejected call without a changed input or resolved blocker.
+Recheck session state when a lifecycle change makes that status useful.
+Make a focused follow-up only when the
+returned result leaves a specific question unanswered.
 
-Complete source Evidence stays accessible through advertised retained references.
-A summary/page has its own Evidence ID; use `normalized_result.parent_evidence_id`
-when a consumer needs the complete source. Pass `{kind: "retained-evidence",
-evidence_id}` in inputs advertising that form; do not resend complete records.
-Sources may independently be inline or retained. Read the Evidence guide for
-exact filters, bundle-based CLI reuse, comparison roles and missing-source recovery.
+If MCP reports `resource_constraint` with `details.resource: "transport"`, call
+`inspect_analysis_view` with its reported same-session Evidence reference for a
+summary, one section/module, or a stable page. Use `trace_application_feature`
+once a seed is known, or call `export_evidence_bundle` to write the complete
+session to a selected path. The analysis remains complete in the ledger; avoid
+repeating it merely to request the same oversized response. Complete CLI JSON
+output also streams without a single MCP frame.
 
-For REA transport refusal or client-truncated output, inspect a smaller retained
-view or export complete Evidence. Do not parse a preview as canonical JSON or
-repeat the producer. Never invent an Evidence ID; if unknown, find it with
-`get_evidence_bundle` summary filters. Retention is bounded and session-local;
-it does not establish automatic persistence or complete source coverage.
+When the retained parent ID is unknown, use `get_evidence_bundle` with
+`detail: "summary"` and exact filters for operation, target SHA-256,
+analysis-profile digest or canonical procedure address. Select its returned
+Evidence ID for the next view instead of requesting the complete bundle. When
+that record has no supported view (for example an annotation), read it with
+`detail: "records"` and an `evidence_id` filter; `missing_evidence_links`
+names linked sources this session does not retain.
+`retention: "complete-record"` means the whole record is stored; native
+value-flow truncation and unknown metadata remain separate. For an exported
+portable bundle, use `rea inspect-evidence-bundle` with JSON containing `path`,
+`detail: "summary"` and the same `filters`. Reading that file still requires
+validating its full contents within the local memory budget.
 
-For broad investigations, read the Evidence guide's staged workflow. Track the
-questions, source boundaries and unresolved gaps; do not claim completion while
-required questions remain open. Inspect packaged resources alongside code when
-relevant, and use runtime evidence only within the requested scope.
+A client can also truncate a successful tool result without a REA transport
+error. Treat that preview as incomplete; use the retained parent ID for smaller
+views or export when the full record is needed. Do not parse the preview as
+canonical JSON or repeat the producer. For JavaScript, read the
+[module-view and trace recipes](references/javascript-applications.md) before
+drilling into a large graph; a broad bidirectional trace can expand unrelated
+containment relationships even after summary analysis.
+
+Every conclusion must distinguish observations, inferences, and unknowns. Cite
+Evidence IDs, preserve limitations and incomplete coverage, and never imply
+that static analysis observed execution. Runtime requests execute the declared
+target and lifecycle; do not broaden the target or action beyond those fields.
+
+Within the user's requested investigation, call available analysis tools
+directly. REA does not require permission grants or per-call approval flags.
+The agent client can independently require approval. Follow the declared request
+scope and the host's actual access requirements; use the recovery guide for an
+observed client-policy rejection.
+
+## Plan broader investigations
+
+For requests that span multiple features or subsystems, use a staged workflow:
+
+1. Turn the request into a checklist of questions and the evidence each answer
+   needs. Resolve target identity and constraints from the conversation and
+   workspace before asking for information again.
+2. Inspect the current REA session, artifact identity, saved analysis database,
+   bookmarks, and prior evidence. Reuse matching state; do not open duplicate
+   sessions or repeat identical analysis.
+3. Start with the smallest useful overview or inventory. Follow each question
+   from its entry point through relevant data and state changes to its result.
+   Batch related operations around a specific hypothesis, then expand only when
+   the returned evidence leaves a concrete gap.
+4. Inspect relevant packaged resources and configuration alongside code when
+   they affect the question. Use format-aware inventory and parsers; do not
+   infer behavior from filenames, strings, or layout alone.
+5. Corroborate a conclusion with the evidence type it requires. Use runtime
+   observation only when static evidence cannot answer the question and the
+   required host runtime and OS access are available. For JavaScript behavior,
+   run probes against the actual app through the available browser, Electron,
+   or process capture workflow.
+6. Decompose work into independent questions. When parallel workers are
+   available and the questions do not depend on one another, assign distinct
+   scopes, point workers to existing evidence, and ask them to return sources,
+   conclusions, and unresolved gaps. Otherwise, work sequentially.
+7. Keep a concise finding ledger linking each conclusion to Evidence IDs,
+   confidence/evidence type, search boundary, and remaining unknowns. Update
+   the shared index or investigation report so later passes can reuse results.
+
+Before finishing, revisit the original checklist. Mark each question as
+answered, partially answered, or unresolved based on its evidence; keep
+bounded negative searches bounded, and do not describe a broad investigation
+as complete while required questions remain open.
+
+## Read only the relevant guide
+
+- Native binaries, managed assemblies, archives, and extraction:
+  [references/native-and-artifacts.md](references/native-and-artifacts.md)
+- ASARs, extracted JavaScript, feature tracing, and version comparison:
+  [references/javascript-applications.md](references/javascript-applications.md)
+- Android APK declarations, classes, methods, and static references:
+  [references/android-applications.md](references/android-applications.md)
+- Native instrumentation, debugger and process capture; browser/Electron
+  observation; static/runtime comparison:
+  [references/runtime-observation.md](references/runtime-observation.md)
+- Evidence paging, comparisons, residual unknowns, and verification:
+  [references/evidence-workflows.md](references/evidence-workflows.md)
+- Agent environments (AgentEnv bundles) of a web application, recorded from
+  a live instance or a HAR:
+  [references/environments.md](references/environments.md)
+- Tasks for a desktop application's exported environment:
+  [references/environments.md](references/environments.md#tasks-for-a-desktop-application-environment)
+- A desktop app environment from a request, built and witnessed on build
+  computers: [references/desktop-build.md](references/desktop-build.md)
 
 ## Finish the task
 
-Explain findings in plain language and tie them to returned Evidence. For builds,
-use normal coding tools and separate observed behavior from design choices.
-Close an opened native/resource session with `close_binary` when no longer needed.
-Evidence references and Unknown history remain on the same MCP connection.
-End with `reset_investigation`; use `save_path` to checkpoint the complete bundle
-before clearing. Import that bundle, including its source closure, to restore
-original references and Unknown history in a new process. Editable Ghidra Program
-checkpoints are separate; read that workflow before edits needing persistence.
+Explain findings in plain language and tie them to returned evidence. When the
+user asks to build something, use normal coding tools and separate observed
+behavior from design choices. Close an opened native session with
+`close_binary` when its engine is no longer needed. Evidence references and Unknown
+history remain in memory on the same MCP connection. End the investigation with
+`reset_investigation`; supply `save_path` to checkpoint the complete bundle before
+clearing. Import the saved bundle to restore original references and Unknown
+history in a new process. Retention is bounded and is not automatic persistence.

.cursor-plugin/skills/reverse-engineer-anything/references/connection-and-recovery.md

@@ -1,23 +1,11 @@
 # Connection and task recovery
 
-Read only the sections needed for the selected task.
-
-- [Choose the remedy from the observed failure](#choose-the-remedy-from-the-observed-failure)
-- [Distinguish client and server deadlines](#distinguish-client-and-server-deadlines)
-- [Recovery ladder](#recovery-ladder)
-- [REA installed as a client plugin](#rea-installed-as-a-client-plugin)
-- [Repair registration when needed](#repair-registration-when-needed)
-- [Codex tool approval is separate from REA readiness](#codex-tool-approval-is-separate-from-rea-readiness)
-- [Repair an editable REA limit](#repair-an-editable-rea-limit)
-- [Continue through an installed CLI](#continue-through-an-installed-cli)
-
 Read this guide only for an observed connection, command, approval or provider
 blocker. Keep the user's target and question in view. Explain what prevented the
 operation and the next useful step; after repair, continue the investigation.
 
 Commands below use an installed `rea`. If none is available, use the package
-invocation under [Continue through an installed CLI](#continue-through-an-installed-cli)
-when acquiring REA is authorized. Follow the connected
+invocation in `SKILL.md` when acquiring REA is authorized. Follow the connected
 server's schemas and the installed CLI's help rather than assuming repository
 main and an older release expose the same arguments. Tool-name prefixes belong
 to the client: find the canonical REA operation in its actual tool listing and
@@ -234,25 +222,6 @@ not edit host deadlines, response buffers, registrations or fixed limits.
 
 ## Continue through an installed CLI
 
-A concrete CLI fallback for an operator-supplied JavaScript tree or ASAR is:
-
-```bash
-npx -y rea-agents@latest analyze-javascript-application /absolute/path/to/app --json
-```
-
-No MCP registration or native engine is required. Read the returned Evidence,
-graph, limitations, and unknowns with the same care as an MCP result; the CLI
-returns the Evidence record directly. This fallback does not establish that
-MCP is configured. Native CLI tasks still require their selected engine.
-Prefer an existing `rea` CLI. When none is installed and REA package acquisition
-is within the authorized scope, use `npx -y rea-agents@latest` in its place from a
-trusted directory outside the target project, with absolute target paths. Managed
-setup binds the current installation directly; retain its runtime and package
-cache entry, or restore that installation before repairing its registration.
-For setup-owned registration, configuration repair starts with a read-only setup plan; show paths, backups and
-skill replacement before applying authorized changes. Preserve existing approval
-and target-execution boundaries.
-
 Check the command environment only if the launcher fails:
 
 ```bash

.cursor-plugin/skills/reverse-engineer-anything/references/environments.md

@@ -1,13 +1,5 @@
 # Agent environments from a live application
 
-Read only the sections needed for the selected task.
-
-- [Before recording](#before-recording)
-- [Record a session an environment can learn writes from](#record-a-session-an-environment-can-learn-writes-from)
-- [Check coverage](#check-coverage)
-- [Export and hand off](#export-and-hand-off)
-- [Tasks for a desktop application environment](#tasks-for-a-desktop-application-environment)
-
 Use this when the user asks for an agent environment (an AgentEnv bundle) of a
 web application: a folder `agent-env` deploys, whose MCP tools answer from what
 one recording of the application observed and serve the writes that recording

.cursor-plugin/skills/reverse-engineer-anything/references/evidence-workflows.md

@@ -1,25 +1,7 @@
 # Evidence, comparison, and verification workflows
 
-Read only the sections needed for the selected task.
-
-- [Read and compose source Evidence](#read-and-compose-source-evidence)
-- [Inventory and decompilation views](#inventory-and-decompilation-views)
-- [Unknowns and comparisons](#unknowns-and-comparisons)
-- [Process captures and environment views](#process-captures-and-environment-views)
-- [Native export table pages](#native-export-table-pages)
-- [Recover oversized delivery and find retained Evidence](#recover-oversized-delivery-and-find-retained-evidence)
-- [Plan broader investigations](#plan-broader-investigations)
-
-## Read and compose source Evidence
-
-Evidence-producing MCP tools return Evidence directly in `structuredContent`.
-CLI complete JSON returns Evidence directly. CLI summary delivery through
-`--evidence_output` returns `{summary, evidence_bundle}`: `summary` is the view's
-Evidence, and `evidence_bundle` is the export receipt with its path and byte count.
-Read `summary.normalized_result.parent_evidence_id` for the complete source ID;
-pass the receipt's path to `--evidence-bundle` on a later supported CLI command.
-Advertised summary delivery retains the complete source separately;
-complete inline delivery is available where the connected contract says so. Read `normalized_result` for operation data,
+Evidence-producing tools return the complete Evidence record directly in MCP
+`structuredContent` and CLI JSON. Read `normalized_result` for operation data,
 and cite `evidence_id` when another tool or your explanation needs a stable
 reference. Preserve authority, limitations, coverage, and residual unknowns. If a result
 is incomplete or paginated, continue only when the remaining data matters to
@@ -67,8 +49,6 @@ An input bundle grants no implicit file lookup: `evidence-file` and historical
 graph file selectors are accepted only by contracts advertising them. See the
 [CLI reuse guide](https://github.com/morluto/rea/blob/main/docs/getting-started/cli.md#import-export-and-compare-evidence).
 
-## Inventory and decompilation views
-
 Completed `inspect_artifact` inventory also supports a summary and
 `view: {kind: "inventory", collection: "occurrences", logical_paths: ["package.json", "main.js"]}`.
 Omit `logical_paths` to page all available occurrences; `[]` selects none.
@@ -101,8 +81,6 @@ exact `where.indices`, `addresses`, `procedure_addresses` or `statuses`. Add
 `collection: "pseudocode"` with one row's `index` to page a large body. Batch
 limitations aggregate every item, so do not attribute them to one successful row.
 
-## Unknowns and comparisons
-
 Use `record_unknown` to track unresolved questions in the user's investigation
 and name the authority or environment still required. Supply supporting and
 contradicting evidence IDs. It updates the session registry directly.
@@ -133,8 +111,6 @@ or absence:
 - `verify_reconstruction` evaluates a finite typed specification. A pass covers
   only declared comparable claims, not global equivalence.
 
-## Process captures and environment views
-
 Process captures are opt-in behavioral evidence, not a security sandbox. V3
 captures cannot be upgraded to V4; rerun the original scenario. Distinguish root
 exit from descendant settlement and require freshness when the task needs it.
@@ -188,8 +164,6 @@ catalog and state size checks. `rea environment-build <capture.har> <dir>`
 inspects, projects and exports in one step and prints the same coverage, or
 `null` when the bundle serves replay.
 
-## Native export table pages
-
 For a PE/ELF export question, `inspect_native_exports` accepts `detail: "summary"`
 and retains the complete table. Query `inspect_analysis_view` with
 `view: {kind: "native-exports", collection: "exports", names: ["selected_name"],
@@ -200,69 +174,3 @@ coverage describes selected retained rows; `summary.coverage` or `exports.covera
 still describes the source table, so an empty partial page does not prove absence.
 CLI `inspect-native-exports --detail summary --evidence_output PATH` saves the
 complete bundle for later `inspect-analysis-view --evidence-bundle PATH` requests.
-
-## Recover oversized delivery and find retained Evidence
-
-If MCP reports `resource_constraint` with `details.resource: "transport"`, call
-`inspect_analysis_view` with its reported same-session Evidence reference for a
-summary, one section/module, or a stable page. Use `trace_application_feature`
-once a seed is known, or call `export_evidence_bundle` to write the complete
-session to a selected path. The analysis remains complete in the ledger; avoid
-repeating it merely to request the same oversized response. Complete CLI JSON
-output also streams without a single MCP frame.
-
-When the retained parent ID is unknown, use `get_evidence_bundle` with
-`detail: "summary"` and exact filters for operation, target SHA-256,
-analysis-profile digest or canonical procedure address. Select its returned
-Evidence ID for the next view instead of requesting the complete bundle. When
-that record has no supported view (for example an annotation), read it with
-`detail: "records"` and an `evidence_id` filter; `missing_evidence_links`
-names linked sources this session does not retain.
-`retention: "complete-record"` means the whole record is stored; native
-value-flow truncation and unknown metadata remain separate. For an exported
-portable bundle, use `rea inspect-evidence-bundle` with JSON containing `path`,
-`detail: "summary"` and the same `filters`. Reading that file still requires
-validating its full contents within the local memory budget.
-
-A client can also truncate a successful tool result without a REA transport
-error. Treat that preview as incomplete; use the retained parent ID for smaller
-views or export when the full record is needed. Do not parse the preview as
-canonical JSON or repeat the producer. For JavaScript, read the
-[module-view and trace recipes](javascript-applications.md) before
-drilling into a large graph; a broad bidirectional trace can expand unrelated
-containment relationships even after summary analysis.
-
-## Plan broader investigations
-
-For requests that span multiple features or subsystems, use a staged workflow:
-
-1. Turn the request into a checklist of questions and the evidence each answer
-   needs. Resolve target identity and constraints from the conversation and
-   workspace before asking for information again.
-2. Inspect the current REA session, artifact identity, saved analysis database,
-   bookmarks, and prior evidence. Reuse matching state; do not open duplicate
-   sessions or repeat identical analysis.
-3. Start with the smallest useful overview or inventory. Follow each question
-   from its entry point through relevant data and state changes to its result.
-   Batch related operations around a specific hypothesis, then expand only when
-   the returned evidence leaves a concrete gap.
-4. Inspect relevant packaged resources and configuration alongside code when
-   they affect the question. Use format-aware inventory and parsers; do not
-   infer behavior from filenames, strings, or layout alone.
-5. Corroborate a conclusion with the evidence type it requires. Use runtime
-   observation only when static evidence cannot answer the question and the
-   required host runtime and OS access are available. For JavaScript behavior,
-   run probes against the actual app through the available browser, Electron,
-   or process capture workflow.
-6. Decompose work into independent questions. When parallel workers are
-   available and the questions do not depend on one another, assign distinct
-   scopes, point workers to existing evidence, and ask them to return sources,
-   conclusions, and unresolved gaps. Otherwise, work sequentially.
-7. Keep a concise finding ledger linking each conclusion to Evidence IDs,
-   confidence/evidence type, search boundary, and remaining unknowns. Update
-   the shared index or investigation report so later passes can reuse results.
-
-Before finishing, revisit the original checklist. Mark each question as
-answered, partially answered, or unresolved based on its evidence; keep
-bounded negative searches bounded, and do not describe a broad investigation
-as complete while required questions remain open.

.cursor-plugin/skills/reverse-engineer-anything/references/firmware-and-raw-images.md

@@ -1,76 +0,0 @@
-# Firmware and raw images
-
-## Firmware regions and extraction
-
-Use `inspect_firmware_regions` when advertised. Use `extract_firmware` when
-extraction is requested, with a caller-selected new absolute output directory.
-These tools use caller-supplied Binwalk/Unblob on Linux; see the
-[firmware guide](https://github.com/morluto/rea/blob/main/docs/guides/firmware/firmware-analysis.md).
-
-## Headerless raw images and firmware containers
-
-For other headerless images (embedded firmware without ELF/Mach-O/PE/MZ
-headers), open with `format: "raw-binary"`, `architecture` (`x86`, `x86_64`,
-`arm`, `arm64`, `6502`, or `m68k`), and `entry_point` (`0x` hex). File offset 0 is mapped at
-that address. Hopper and IDA refuse this interpretation; Ghidra uses
-`BinaryLoader` with a generic language: big-endian for `m68k`, little-endian
-otherwise. Big-endian ARM, Thumb-only, and Cortex-M variants are unsupported.
-`6502` selects NMOS 6502; the entry and complete file mapping must fit within
-`0x0000..0xffff`. 65C02, 65816, cartridge containers, bank switching, and mapper
-semantics are unsupported. `m68k` requires an even entry and decodes the 68040
-superset of the 68000 family; Amiga/Atari containers and vector-table entry
-discovery are unsupported. Windows Ghidra P0 admits raw images only for `6502`. See [raw binary guide](https://github.com/morluto/rea/blob/main/docs/providers/raw-binary.md).
-
-Intel HEX, S-record, UF2 and DfuSe files are containers, not raw images:
-`inspect_artifact` lists address regions (UF2 per familyID, DfuSe per element
-with its alternate setting), and `extract_artifact` writes each region. Open an
-extracted region as `raw-binary` with its start address as `entry_point`. UF2
-numbering gaps leave coverage unknown; a DfuSe suffix CRC mismatch follows
-`integrity_policy`. See [UF2 and DfuSe](https://github.com/morluto/rea/blob/main/docs/uf2-dfuse.md).
-
-## ESP images
-
-For an ESP32-family app or bootloader image, `open_binary` then
-`inspect_artifact` with `detail: "summary"`. `root_format_metadata` holds the
-observed `header` and `descriptor` (its `kind` is the app or bootloader role)
-and the derived `chip`, `checksum`, `appended_sha256` and
-`undescribed_trailing` ranges. Segment rows (`segment-NN.bin`) carry
-`load_addr`, `classification` (`mapped`, `loaded`, `not-loaded` or `unknown`,
-per ESP-IDF `should_map`/`should_load`; RAM ranges are not checked) and
-`source_location`; select them with `inspect_analysis_view` and
-`logical_paths`. Signature blocks are `unverified`; `public_key_sha256`
-matches eFuse key digests.
-
-Unknown `source_coverage` means truncation or undescribed trailing bytes, such
-as the flash segments of a `--ram-only-header` image; do not infer absence
-from it. Erased flash is described. Integrity mismatches are facts, not
-refusals. ESP8266 images are refused, and flash-encrypted images are not
-recognized. Segments are not disassembled, and RISC-V or Xtensa segments
-cannot be opened as raw-binary. See
-[ESP images](https://github.com/morluto/rea/blob/main/docs/esp-image.md).
-
-## ESP flash images
-
-A whole-flash dump opens as `esp-flash-image`. `root_format_metadata` holds
-`partition_table` (its `method`: `default`, `caller` or a derived
-`scan`), `bootloader` and `apps` ESP image facts, the `otadata` records,
-`partitions_outside_dump` and `boot_selection`. Every table has a
-`boot_selection`: its `basis` names the bootloader's branch (`no-otadata`
-without otadata, `none` when the bootloader rejects the table, `unknown`
-when a record is missing), and it lists its `assumptions`;
-`running_partition` is always `unknown`. Each image names its entry
-`path`. With `detail: "summary"` the root kind is `esp-flash-image-summary`
-and each image is a compact `summary` (chip, entry point, segment count,
-classified load ranges, signature block count, descriptor identity,
-integrity); complete detail keeps every segment. Partition rows
-(`partitions/<label>.bin`) carry the declared `type`, `subtype`, names,
-`flash_address`, `size` and flags; `encrypted` is a storage flag, not
-proof of ciphertext. `written_offset`/`written_end` bound the bytes that
-are not erased: an app slot whose facts say `erased` can still hold an older
-image, and a limitation says so. For a dump read from another flash address, call
-`open_binary` with `format: "esp-flash-image"` and `flash_offset`; an
-`esp-image` whose `flash_dump` is set names that offset in a next probe; a
-relocated table can be named with `partition_table_offset`. No table found
-means unknown, often flash encryption. Extraction writes `nvs_keys` as
-ordinary data. See
-[ESP flash images](https://github.com/morluto/rea/blob/main/docs/esp-flash-image.md).

.cursor-plugin/skills/reverse-engineer-anything/references/format-inspectors.md

@@ -1,449 +0,0 @@
-# Offline format and metadata inspection
-
-Read only the sections needed for the selected task.
-
-- [Release exposure](#release-exposure)
-- [Go build metadata](#go-build-metadata)
-- [Model files](#model-files)
-- [IL2CPP metadata](#il2cpp-metadata)
-- [Rust dependency metadata](#rust-dependency-metadata)
-- [PE Authenticode certificate tables](#pe-authenticode-certificate-tables)
-- [PE import and delay-load tables](#pe-import-and-delay-load-tables)
-- [Windows Shell Link shortcuts](#windows-shell-link-shortcuts)
-- [PE TLS callbacks and load configuration](#pe-tls-callbacks-and-load-configuration)
-- [Binary hardening](#binary-hardening)
-- [ELF note tables](#elf-note-tables)
-- [ELF and Mach-O initializer functions](#elf-and-mach-o-initializer-functions)
-- [ELF unwind function ranges](#elf-unwind-function-ranges)
-- [Apple IPS crash reports](#apple-ips-crash-reports)
-- [Linux kernel modules](#linux-kernel-modules)
-- [Linux kernel build config](#linux-kernel-build-config)
-- [CPython pyc files](#cpython-pyc-files)
-- [Erlang BEAM files](#erlang-beam-files)
-- [SQLite snapshots](#sqlite-snapshots)
-- [Recorded Linux crashes](#recorded-linux-crashes)
-- [Offline ELF layout](#offline-elf-layout)
-- [Offline WASM](#offline-wasm)
-- [Offline APK signatures](#offline-apk-signatures)
-- [Offline PE resource inventory](#offline-pe-resource-inventory)
-- [Offline PE and ELF export tables](#offline-pe-and-elf-export-tables)
-- [EVM bytecode](#evm-bytecode)
-- [Embedded GPU code objects](#embedded-gpu-code-objects)
-
-## Release exposure
-
-Use `audit_release_exposure` (`rea audit-release-exposure PATH`) when the question is what static analysis of a shipped `.app`, install directory, ASAR, or executable recovers. Each check has a stable id, a status, the observation behind it, and fixed remediation. Unknown and unassessed checks stay in the report. The report does not establish that the application is secure.
-
-`text.credential_markers` reports a rule id and location only. Matched bytes, previews, lengths, and hashes of matched bytes are omitted.
-
-`native.hardening`, `macho.hardened_runtime`, and `android.manifest` are not assessed here. Their next probes are `inspect_binary_hardening`, `inspect_signature`, and `inspect_android_package`.
-
-See [release exposure](https://github.com/morluto/rea/blob/main/docs/release-exposure.md).
-
-## Go build metadata
-
-Use `inspect_go_binary` for an explicit local ELF, PE or Mach-O (universal
-files need `architecture`) when the
-question concerns the embedded compiler version, module/dependency versions,
-local replacements, build settings, or static Go string literals packed without
-NUL terminators. Pass `include_string_literals: true`
-(`--include-string-literals`, optional `max_string_literals` /
-`--max-string-literals`) to recover bounded inferred word-aligned `(ptr, len)`
-UTF-8 string slices from read-only data and page them through `go`
-`string_literals` views. No `open_binary`, Go installation or
-target execution is required. The CLI equivalent is `inspect-go-binary PATH`.
-Preserve artifact SHA-256, original file ranges, framed module bytes and every
-unparsed module-text line. A missing record is an unknown, not proof of a
-non-Go binary; observed metadata does not authenticate its claims, and static
-`(ptr, len)` string recovery excludes compiler-inlined stack/immediate strings
-without proving runtime reachability. Function,
-source and type recovery remain a separate native-analysis question.
-
-See the [Go binary guide](https://github.com/morluto/rea/blob/main/docs/guides/native/go-binaries.md).
-
-## Model files
-
-Optional `view` selects one page in the initial call. Batch metadata keys or
-tensor names and select only needed fields; do not infer model behavior from
-declared architecture or tensor types.
-
-Use `inspect_model_artifact` (`inspect-model-artifact PATH`) for local GGUF v2/v3
-and safetensors declarations without a model runtime. Header-only mode does not
-read weights or establish full-file identity. `hash: "full"` explicitly streams
-all bytes. Retained `model-artifact` views page `metadata`, `tensors` and derived
-`types`, with exact batch filters and selected fields. Arrays retain typed counts
-and eight-element previews; omitted values, unknown types and runtime behavior
-stay explicit. See [model files](https://github.com/morluto/rea/blob/main/docs/model-artifacts.md).
-
-## IL2CPP metadata
-
-Use `inspect_il2cpp_metadata` for an explicit local Unity `global-metadata.dat`
-(versions 24, 27, 29, 31 and 39) when the question concerns assembly images,
-names, member owners, string literals, namespaces or index ranges. No `open_binary`, Unity installation or target
-execution is required. The CLI equivalent is `inspect-il2cpp-metadata PATH`.
-MCP defaults to counts with complete retained Evidence. Query `sections`,
-`images`, `types`, `methods`, `fields`, `properties`, `events`, `parameters` or
-`string_literals` through `inspect_analysis_view`; use exact names,
-batch indices and selected fields. Row indices are scoped by artifact SHA-256
-and collection; page offsets count filtered matches. Parameters support batch
-`method_indices` to read one method's parameters directly. Historical reports
-without member collections retain unknown counts rather than empty inventories. Preserve
-artifact SHA-256 and index widths. Unrecognized magic may mean encryption and
-is unsupported, not decrypted.
-
-Optional matching `binary_path` enables static native addresses and type
-references. Query `method_addresses` with indices or addresses/states, preserving
-every shared-body method and both artifact identities. Without a linked binary,
-native counts remain unknown. Byte-pattern fallback is inferred; no runtime
-behavior or same-build pairing is proved.
-See [IL2CPP metadata](https://github.com/morluto/rea/blob/main/docs/il2cpp-metadata.md).
-
-## Rust dependency metadata
-
-For selected crates, use `view: {collection: "packages", names: [...],
-fields: ["name", "version", "source"]}` on the first call. Include `dependencies`
-explicitly for direct index references. Select `detail: "summary"` for counts and
-ranges; retained Rust package/edge views support later filters. Keep complete
-parent Evidence accessible; CLI compact delivery requires `--evidence_output`.
-
-Use `inspect_rust_binary` for an explicit local ELF, PE, or Mach-O (universal
-files need `architecture`) when
-the question concerns the cargo-auditable crate list embedded in `.dep-v0`
-(Mach-O: `__DATA,.dep-v0`). No `open_binary`, Rust installation, or target
-execution is required. The CLI equivalent is `inspect-rust-binary PATH`.
-Preserve artifact SHA-256, package order, dependency indices, and the section
-file range. A missing section is null metadata, not proof the binary has no
-dependencies. An omitted package kind stays null. A rustc commit hash is
-inferred from a read-only `/rustc/<40-hex>/` path and does not by itself
-identify the compiler. See [Rust binaries](https://github.com/morluto/rea/blob/main/docs/rust-binaries.md).
-
-## PE Authenticode certificate tables
-
-Use `inspect_pe_certificates` for an explicit local PE32/PE32+ image when the
-question concerns the Authenticode certificate table, WIN_CERTIFICATE types, or
-embedded PKCS#7 signer and X.509 identity fields. No `open_binary`, Windows SDK,
-or target execution is required. The CLI equivalent is
-`inspect-pe-certificates PATH`. Preserve artifact SHA-256, table file ranges,
-payload hashes, original validity-time spellings, and
-`subject_key_identifier_hex` when the signer uses that form. A missing
-`IMAGE_DIRECTORY_ENTRY_SECURITY` directory is observed absence of that table,
-not proof the image is unsigned in every catalog or overlay sense. Signatures,
-chains, timestamps, and Authenticode digest coverage remain unobserved. See
-[PE Authenticode certificate tables](https://github.com/morluto/rea/blob/main/docs/pe-certificates.md).
-
-## PE import and delay-load tables
-
-Use `inspect_pe_imports` for an explicit local PE32/PE32+ image when the
-question concerns imported DLL and function names, ordinals, delay-load
-descriptors, or the pefile import hash. No `open_binary`, Windows SDK, or
-target execution is required. The CLI equivalent is `inspect-pe-imports PATH`.
-Preserve artifact SHA-256, directory file ranges, original DLL spellings, and
-ILT/IAT RVAs. A missing `IMAGE_DIRECTORY_ENTRY_IMPORT` directory is observed
-absence of that table, not proof the image never imported APIs through another
-table or overlay. Bound snapshots, loaded IAT contents, and export-forward
-resolution remain unobserved. See
-[PE import and delay-load tables](https://github.com/morluto/rea/blob/main/docs/pe-imports.md).
-
-## Windows Shell Link shortcuts
-
-Use `inspect_windows_shortcut` for an explicit local `.lnk` file when the
-question concerns the recorded target path, arguments, working directory, icon
-location, volume serial, network share names, ExtraData blocks, or Tracker
-droid identifiers. No `open_binary`, Windows SDK, COM, or target execution is
-required. The CLI equivalent is `inspect-windows-shortcut PATH`. Preserve
-artifact SHA-256, FILETIME raw decimals, IDList type bytes, and ExtraData
-signatures. Unknown IDList items and ExtraData signatures remain raw with
-SHA-256; they are not guessed. Version-1 UUID MAC and timestamp values are
-labelled derived. A missing optional section is observed absence of that
-structure. See
-[Windows Shell Link shortcuts](https://github.com/morluto/rea/blob/main/docs/guides/native/windows-shortcut.md).
-
-## PE TLS callbacks and load configuration
-
-Preserve each callback's RVA, file offset and section alongside its VA and the
-size-covered declaration fields. A missing directory is observed absence of
-that declaration.
-
-Use `inspect_pe_load_config` for an explicit local PE32/PE32+ image when the
-question concerns code that runs before the entry point through TLS callbacks,
-or the security cookie, SafeSEH, CFG, EH-continuation and XFG declarations in
-the load configuration. No `open_binary`, Windows SDK, or target execution is
-required. The CLI equivalent is `inspect-pe-load-config PATH`. Preserve artifact
-SHA-256, directory and callback-array file ranges, and callback VAs as
-hexadecimal strings. Fields outside the declared `Size` are `null`. Guard-flag
-mitigations carry `evidence_kind: "inferred"`; enforcement remains unobserved.
-See
-[PE TLS callbacks and load configuration](https://github.com/morluto/rea/blob/main/docs/pe-load-config.md).
-
-## Binary hardening
-
-Each property reports `enabled`, `disabled`, `partial`, `unknown`, or
-`not_applicable`, with the header field or import behind it.
-
-Use `inspect_binary_hardening` for an explicit local ELF, PE, or Mach-O file
-when the question concerns PIE or ASLR, NX, RELRO, stack-protector or FORTIFY
-signs, control-flow flags, or stripped state. No `open_binary`, pwntools, or
-target execution is required. The CLI equivalent is
-`inspect-binary-hardening PATH`. Universal Mach-O files return one image per
-slice. A missing import is `unknown`, not proof the protection was omitted.
-Authenticode and Mach-O code-signature rows report presence, not validity.
-`hardened_runtime` stays unknown and names `inspect_signature`. See
-[binary hardening](https://github.com/morluto/rea/blob/main/docs/binary-hardening.md).
-
-## ELF note tables
-
-Retain each note's owner, type, descriptor SHA-256, and PT_NOTE/SHT_NOTE ranges.
-
-Use `inspect_elf_notes` for an explicit local ELF32/ELF64 image when the
-question concerns PT_NOTE/SHT_NOTE owners, GNU build-id, or GNU ABI-tag
-fields. No `open_binary` or target execution is required. The CLI equivalent
-is `inspect-elf-notes PATH`. Preserve artifact SHA-256, table file ranges,
-descriptor hashes, and original owner spellings. A missing table is observed
-absence of those declarations, not proof the image never carried notes in
-another container. See
-[ELF note tables](https://github.com/morluto/rea/blob/main/docs/elf-notes.md).
-
-## ELF and Mach-O initializer functions
-
-Use `inspect_native_initializers` for an explicit local ELF or Mach-O image
-when the question is which functions run before `main` or at exit: ELF
-`DT_PREINIT_ARRAY`, `DT_INIT`, `DT_INIT_ARRAY`, `DT_FINI`, `DT_FINI_ARRAY` and
-legacy `.ctors`/`.dtors`, or Mach-O `__mod_init_func`, `__init_offsets` and
-`__mod_term_func`. Pass `architecture` for a universal Mach-O. The CLI
-equivalent is `inspect-native-initializers PATH`. PIE slots are resolved
-through RELATIVE, RELR and section relocations, and Mach-O slots through
-chained fixups. Keep each entry's `resolution`: `unresolved` means the value was
-not established, and `unresolved_import` names an external symbol. Page with
-`offset` and `limit`. See
-[ELF and Mach-O initializer functions](https://github.com/morluto/rea/blob/main/docs/guides/native/native-initializers.md).
-
-## ELF unwind function ranges
-
-Use `inspect_elf_unwind_ranges` for an explicit local ELF32/ELF64 image when
-the question concerns function starts and lengths that survive `strip`, which
-functions have an LSDA, or which ranges have no exact symbol. No
-`open_binary` or target execution is required. The CLI equivalent is
-`inspect-elf-unwind-ranges PATH`. Preserve artifact SHA-256, FDE start and
-end, containing section, exact symbol name, CIE index, LSDA, and personality.
-Page later ranges with `offset` and `limit`. A missing `.eh_frame` is
-observed absence of that declaration and of `PT_GNU_EH_FRAME`, not proof the
-image has no functions. Header disagreement is a limitation and does not drop
-decoded ranges. `.debug_frame` is reported present or absent and is not
-decoded. CFA instructions are counted and not evaluated. Indirect EH pointers
-are not dereferenced. See
-[ELF unwind function ranges](https://github.com/morluto/rea/blob/main/docs/guides/native/elf-unwind-ranges.md).
-
-## Apple IPS crash reports
-
-Use `inspect_apple_ips` for an explicit local Apple `.ips` file when the
-question concerns crash metadata, exception and termination facts, threads,
-usedImages, lossless registers, or documented crash classes. No `open_binary`,
-Xcode, atos, or target execution is required. The CLI equivalent is
-`inspect-apple-ips PATH`. Preserve artifact SHA-256, `bug_type` as a string,
-uppercase image UUIDs, and integer literals that exceed IEEE-754 range. Only
-`bug_type` `"309"` yields a crash body; other DiagnosticReports types return
-no crash result. Select `join_path` for one local Mach-O or dSYM to match
-report images to slices by UUID and architecture. Preserve the selected
-artifact identity, slice facts and unjoined reasons; a joined static address
-does not resolve function names or dSYM symbols. See
-[Apple IPS crash reports](https://github.com/morluto/rea/blob/main/docs/apple-ips.md).
-
-## Linux kernel modules
-
-Use `inspect_kernel_module` for an explicit local ELF32/ELF64 relocatable
-object when the question concerns `.modinfo` license, vermagic, alias,
-depends, or parm strings, the `__versions` symbol CRCs, or an appended module
-signature trailer. No `open_binary`, Linux host, or `modinfo` program is
-required. The CLI equivalent is `inspect-kernel-module PATH`. Preserve
-artifact SHA-256, key order, duplicate keys, CRC hex, and section file ranges.
-A missing section is observed absence when section names are known, not proof
-the module never carried that metadata. `.gnu.linkonce.this_module` is reported
-as present or absent; `struct module` stays unknown because its layout changes
-with the kernel version. `__ksymtab` records are not decoded. The PKCS#7 range
-and its SHA-256 are observed; certificate, signer, and digest coverage remain
-unvalidated. An absent trailer is not proof the module is unsigned in another
-container. gzip, xz, and zstd whole-file containers are unsupported and name
-`external_decompression` as the next probe: use an external codec utility to
-decompress into a separate file, then call `inspect_kernel_module` on that file. See
-[Linux kernel modules](https://github.com/morluto/rea/blob/main/docs/kernel-modules.md).
-
-## Linux kernel build config
-
-Use `inspect_kernel_build_config` for an explicit local kernel image (`vmlinux`
-ELF, raw arm64 `Image`, x86 `bzImage`/`zImage`, or `configs.ko`) when the
-question is which kernel version it is or how it was configured. No
-`open_binary`, Linux host, or `extract-ikconfig` is required. The CLI
-equivalent is `inspect-kernel-build-config PATH`; pass `name` for one exact
-`CONFIG_*` option and `offset`/`limit` to page the rest. Preserve the banner
-version, builder, and compiler, the config SHA-256 and option count, and each
-entry's value kind (`y`, `m`, `string`, `number`, `not_set`). A config is
-accepted only when `IKCFG_ST` is immediately followed by a valid gzip member
-and `IKCFG_ED`, so an ordinary `IKCFG_ST` string is not a config. Offsets with
-`space: "inflated_payload"` are positions in an inflated bzImage payload, not
-file offsets. An absent config describes only the searched bytes; the kernel
-may have been built without `CONFIG_IKCONFIG` or with it as a module. Truncated
-payloads are incomplete, not absent. A framed blob that does not decode, has a
-bad CRC, or holds non-config text is `malformed`, not absent. xz, lz4, lzma, zstd, bzip2, and lzo
-payloads are unsupported and name `external_decompression`: decompress with an
-external utility, then call `inspect_kernel_build_config` on the result. See
-[Linux kernel build config](https://github.com/morluto/rea/blob/main/docs/guides/native/kernel-build-config.md).
-
-## CPython pyc files
-
-Use `inspect_python_pyc` for an explicit local CPython 3.7–3.13 `.pyc` when the
-question concerns the header magic, PEP 552 source hash or timestamp, nested
-code-object names, or bytecode identity. No `open_binary`, Python installation,
-or target execution is required. The CLI equivalent is `inspect-python-pyc PATH`.
-Preserve artifact SHA-256, the 16-byte header fields, nested code names, and
-bytecode SHA-256. Bytecode is not decompiled. A decoded header does not prove
-the matching `.py` is present. See [CPython pyc inspection](https://github.com/morluto/rea/blob/main/docs/python-pyc.md).
-
-## Erlang BEAM files
-
-Use `inspect_erlang_beam` for an explicit local `FOR1`/`BEAM` file when the
-question concerns IFF chunks, atom names, exports, imports, lambdas, or the
-Code header. No `open_binary`, Erlang installation, or bytecode execution is
-required. The CLI equivalent is `inspect-erlang-beam PATH`. Preserve artifact
-SHA-256, chunk file ranges, 1-based atom indices, and original atom spellings.
-Missing optional tables stay null. For a few facts, use `detail: "summary"` with
-a `beam` view selecting a collection, exact names/indices and fields; read later
-pages from `parent_evidence_id` or the complete Evidence bundle. Same-name
-exports retain every arity and original atom index. CLI summary delivery requires
-`--evidence_output`. Null table counts remain distinct from decoded empty tables. See
-[Erlang BEAM files](https://github.com/morluto/rea/blob/main/docs/erlang-beam.md).
-
-## SQLite snapshots
-
-Use `inspect_sqlite_database` with an explicit local database snapshot for its
-schema and optional rows from one exact ordinary table. It requires Node.js
-24.x >=24.15 or 26+ with the native `DatabaseSync.limits` API. It also consumes the
-existing sibling `-wal`; retain both original digests when citing its result.
-Default output is a compact inventory; a selected table expands only its own
-schema and samples 20 rows. Select `columns` to avoid irrelevant or large cells,
-and request `schema_scope: "database"` only for complete schema details.
-Choose `row_limit` when a finite sample is useful and keep truncation and
-unspecified row order explicit. Virtual tables, shadow tables and views are
-schema facts, not row-read targets. REA inspects private copies and leaves
-the source files untouched; obtain a coherent quiescent database/WAL snapshot
-before inspection. See [SQLite snapshots](https://github.com/morluto/rea/blob/main/docs/guides/data/sqlite-databases.md).
-
-## Recorded Linux crashes
-
-Use `inspect_recorded_crash` with an explicit Linux x86-64 ELF core path to
-inspect historical thread registers, signals, raw notes and original file
-ranges. It is independent of the active disassembler target. Source-bound
-register values and note bytes are observations; missing notes, signal-thread
-association and current executable/library identity remain unknown. PIDs are
-historical metadata and never authorize live attach or control.
-
-The optional `include_debugger_context` facet adds mapping candidates through
-caller-supplied GDB/pwndbg in an owned core-only session. Display names do not
-establish current file identity; zero reported flags leave permissions unknown.
-Requested unavailable context fails with setup guidance. See
-[recorded crashes](https://github.com/morluto/rea/blob/main/docs/guides/native/recorded-crashes.md)
-for exact upstream profiles, bounds and verification coverage.
-
-## Offline ELF layout
-
-Use `inspect_binary_layout` with an explicit local ELF path for file-backed
-sections, symbols, relocations, linkage names, and static mitigation inferences
-without opening a disassembler target. When that complete record is already
-retained, call `inspect_analysis_view` for a summary, mitigations or linkage
-facet, one section or symbol, or a stable page. Do not re-run the decoder to
-read a single object.
-
-## Offline WASM
-
-Supply WABT 1.0.42 through `REA_WABT_BIN_DIRECTORY`. Select JavaScript glue and
-candidate paths explicitly, preserving URL/path ambiguity and exact byte/tool
-identity; no implicit fetch is performed.
-
-Use `inspect_wasm_artifact` with caller-supplied WABT for selected local bytes.
-For counts, validation and WAT identity, select `detail: "summary"` on the first
-MCP call. The complete original Evidence stays retained; use
-`normalized_result.parent_evidence_id` for exact exports or text pages. Omitted
-or complete detail keeps the existing complete result.
-For a completed result, use `inspect_analysis_view` with its reported retained
-reference for a summary or `view: {kind: "wasm", facet: "exports", offset: 0, limit: 32}`.
-Facets also include sections, imports, candidates, glue, WAT, headers, details
-and tool profile. Text offsets count UTF-16 code units; follow `next_offset`.
-The complete source remains retained after an oversized MCP reply. No target
-execution or network acquisition is implied, and WAT is decoded representation.
-
-For exact core-module import/export differences, use `compare_wasm_interfaces`
-with two complete inspection Evidence records (or current-session retained
-references) and explicit scope. Preserve unknown types/duplicate names; static
-interface equality is not runtime compatibility. CLI accepts complete inline
-Evidence in `rea compare-wasm-interfaces input.json --json`, or exact references
-with `--evidence-bundle /absolute/path/evidence.json`. Use the complete parent
-ID when the inspection returned a summary.
-
-See the [WASM guide](https://github.com/morluto/rea/blob/main/docs/guides/wasm/wasm-artifacts.md).
-
-## Offline APK signatures
-
-Use `inspect_apk_signature` for a selected local APK/ZIP/AAB. Default MCP summaries
-retain complete Evidence. Select `view.collection: signers/digests/lineage`,
-with scheme or ID batches, fields and paging; use `inspect_analysis_view` kind
-`apk-signature` for retained reads after deletion. IDs retain artifact/scheme
-scope and signer links; `fields: []` keeps identity links. Empty pages preserve
-partial decoding reasons and unknown authenticity. CLI compact delivery needs
-`--evidence_output` to preserve the complete parent.
-It reads the APK
-Signing Block (v2/v3/v3.1) and v1 `META-INF` signing without executing anything,
-and reports each signer's certificate facts (for v1, the certificate the PKCS#7
-block's SignerInfo names), the declared v3 rotation lineage,
-the declared source-stamp certificate and, for SHA-256/SHA-512 digests, whether
-the declared content digest matches the current bytes. That match is not a
-signature verification: anyone who changes the bytes can recompute the digest.
-`posture` summarises v1-only signing (which cannot tell a v1-only APK from one
-whose v2/v3 block was removed), observed signer agreement across schemes,
-whether the v1 signer equals the block-scheme signer (`false` is notable),
-whether every differing signer appears in a declared (unverified) rotation
-lineage, and the strongest scheme.
-It never claims authenticity: no signature or lineage is cryptographically
-verified.
-
-See the [APK signature guide](https://github.com/morluto/rea/blob/main/docs/apk-signature.md).
-
-## Offline PE resource inventory
-
-Use `inspect_pe_resources` with an absolute `path`, or
-`rea inspect-pe-resources ./selected.exe --json`, for portable PE32/PE32+
-resource inspection without opening a native target. Preserve returned Evidence:
-type/name/language IDs and UTF-16 names, original directory/data/payload ranges,
-code pages, payload hashes and static icon-group references. Same-language icon
-resolution is explicit; missing references and size mismatch remain visible.
-Other payloads are opaque, and this operation does not decode images or simulate
-Windows resource loading. Consult the budgets and supported tree profile in
-[offline binary diagnostics](https://github.com/morluto/rea/blob/main/docs/guides/native/binary-diagnostics.md).
-
-## Offline PE and ELF export tables
-
-Use `inspect_native_exports` with an absolute `path`, or
-`rea inspect-native-exports ./libexample.so --json`, to read a PE export
-directory or ELF dynamic symbol table on any host without opening a native
-target. Ordinal-only, forwarded and hidden-version exports stay distinct;
-partial or unsupported coverage means the inventory is incomplete, not that
-exports are absent. Use `inspect_macho` for Mach-O. See
-[offline binary diagnostics](https://github.com/morluto/rea/blob/main/docs/guides/native/binary-diagnostics.md).
-
-## EVM bytecode
-
-Explicit local EVM bytecode carrier: `inspect_evm_interface` with caller-selected
-`raw` or `hex` encoding. The owned worker deadline is `timeout_ms` (default
-30000 ms, range 1..2147483647, next call). Select byte, address-space and CPU
-allowances separately through `producer_limits`; tighter inherited OS limits
-remain enforced, and the 256 MiB worker heap is independent. Preserve carrier/decoded digests and treat selectors,
-argument strings and mutability as inferred candidates; this performs no chain
-lookup or target execution. For the complete inferred interface without hex or
-duplicate stdout, discover support and select `detail: "summary"`; complete
-Evidence stays retained under its original identity. Use the returned
-`parent_evidence_id` for exact function or `bytecode_hex`/stdout/stderr pages.
-Candidate coverage does not establish interface completeness. See the
-[EVM bytecode guide](https://github.com/morluto/rea/blob/main/docs/guides/evm/evm-bytecode.md).
-
-## Embedded GPU code objects
-
-Embedded CUDA/HIP code objects: `inspect_gpu_code_objects`; inventory targets
-first, then decode selected payloads and query retained GPU Evidence.

.cursor-plugin/skills/reverse-engineer-anything/references/javascript-applications.md

@@ -1,13 +1,5 @@
 # JavaScript and Electron application artifacts
 
-Read only the sections needed for the selected task.
-
-- [Inventory route candidates without returning the application graph](#inventory-route-candidates-without-returning-the-application-graph)
-- [Read a module without repeating analysis](#read-a-module-without-repeating-analysis)
-- [Trace only the remaining question](#trace-only-the-remaining-question)
-- [Recover an oversized display](#recover-an-oversized-display)
-- [Reusing application Evidence](#reusing-application-evidence)
-
 For captured or bundled JavaScript that needs readable source recovery, use
 `recover_javascript_sources` with an absolute `path` and absent
 `output_directory`. Supply related runtime/shared/lazy files in
@@ -98,9 +90,6 @@ Endpoint candidates remain inferences rather than confirmed network behavior.
 ## Read a module without repeating analysis
 
 Use the actual returned parent ID, on the same connection, as `source.evidence_id`.
-For CLI summary exports, take that ID from
-`summary.normalized_result.parent_evidence_id` and pass `--evidence-bundle` with
-the exported bundle's path to each later `inspect-analysis-view` invocation.
 The following IDs are placeholders to replace with returned values. When the
 module's node ID is not already known, page identities before selecting an item.
 Pages return only node IDs, paths and roles; choose a useful page size and follow

.cursor-plugin/skills/reverse-engineer-anything/references/native-and-artifacts.md

@@ -1,23 +1,4 @@
-# Native and managed analysis
-
-<a id="native-managed-and-packaged-artifacts"></a>
-
-Read only the sections needed for the selected task.
-
-- [Other artifact guides](#other-artifact-guides)
-- [Apple app resources](#apple-app-resources)
-- [Desktop installation directories](#desktop-installation-directories)
-- [Native targets](#native-targets)
-- [Continue editable Ghidra analysis across processes](#continue-editable-ghidra-analysis-across-processes)
-- [Reuse caller-selected Ghidra analysis](#reuse-caller-selected-ghidra-analysis)
-- [Ghidra startup deadline recovery](#ghidra-startup-deadline-recovery)
-- [Bridge request ceiling](#bridge-request-ceiling)
-- [Selected native facts and call graphs](#selected-native-facts-and-call-graphs)
-- [Static memory and original file bytes](#static-memory-and-original-file-bytes)
-- [DOS MZ with Ghidra](#dos-mz-with-ghidra)
-- [Managed PE/CLI](#managed-pe-cli-guide)
-- [DOS COM loader interpretation](#dos-com-loader-interpretation)
-- [Function annotations in Ghidra](#function-annotations-in-ghidra)
+# Native, managed, and packaged artifacts
 
 For release comparisons on one MCP connection, retain `list_procedures` Evidence
 from each target and reuse its exact IDs after target switches or `close_binary`.
@@ -81,39 +62,342 @@ runtime presence. For compact CLI output, use
 `inspect-desktop-application PATH --detail summary --evidence_output BUNDLE`.
 See [desktop installations](https://github.com/morluto/rea/blob/main/docs/desktop-applications.md).
 
+## Release exposure
+
+Use `audit_release_exposure` (`rea audit-release-exposure PATH`) when the question is what static analysis of a shipped `.app`, install directory, ASAR, or executable recovers. Each check has a stable id, a status, the observation behind it, and fixed remediation. Unknown and unassessed checks stay in the report. The report does not establish that the application is secure.
+
+`text.credential_markers` reports a rule id and location only. Matched bytes, previews, lengths, and hashes of matched bytes are omitted.
+
+`native.hardening`, `macho.hardened_runtime`, and `android.manifest` are not assessed here. Their next probes are `inspect_binary_hardening`, `inspect_signature`, and `inspect_android_package`.
+
+See [release exposure](https://github.com/morluto/rea/blob/main/docs/release-exposure.md).
+
+## Go build metadata
+
+Use `inspect_go_binary` for an explicit local ELF, PE or Mach-O (universal
+files need `architecture`) when the
+question concerns the embedded compiler version, module/dependency versions,
+local replacements, build settings, or static Go string literals packed without
+NUL terminators. Pass `include_string_literals: true`
+(`--include-string-literals`, optional `max_string_literals` /
+`--max-string-literals`) to recover bounded inferred word-aligned `(ptr, len)`
+UTF-8 string slices from read-only data and page them through `go`
+`string_literals` views. No `open_binary`, Go installation or
+target execution is required. The CLI equivalent is `inspect-go-binary PATH`.
+Preserve artifact SHA-256, original file ranges, framed module bytes and every
+unparsed module-text line. A missing record is an unknown, not proof of a
+non-Go binary; observed metadata does not authenticate its claims, and static
+`(ptr, len)` string recovery excludes compiler-inlined stack/immediate strings
+without proving runtime reachability. Function,
+source and type recovery remain a separate native-analysis question.
+
+## Model files
+
+Use `inspect_model_artifact` (`inspect-model-artifact PATH`) for local GGUF v2/v3
+and safetensors declarations without a model runtime. Header-only mode does not
+read weights or establish full-file identity. `hash: "full"` explicitly streams
+all bytes. Retained `model-artifact` views page `metadata`, `tensors` and derived
+`types`, with exact batch filters and selected fields. Arrays retain typed counts
+and eight-element previews; omitted values, unknown types and runtime behavior
+stay explicit. See [model files](https://github.com/morluto/rea/blob/main/docs/model-artifacts.md).
+
+## IL2CPP metadata
+
+Use `inspect_il2cpp_metadata` for an explicit local Unity `global-metadata.dat`
+(versions 24, 27, 29, 31 and 39) when the question concerns assembly images,
+names, member owners, string literals, namespaces or index ranges. No `open_binary`, Unity installation or target
+execution is required. The CLI equivalent is `inspect-il2cpp-metadata PATH`.
+MCP defaults to counts with complete retained Evidence. Query `sections`,
+`images`, `types`, `methods`, `fields`, `properties`, `events`, `parameters` or
+`string_literals` through `inspect_analysis_view`; use exact names,
+batch indices and selected fields. Row indices are scoped by artifact SHA-256
+and collection; page offsets count filtered matches. Parameters support batch
+`method_indices` to read one method's parameters directly. Historical reports
+without member collections retain unknown counts rather than empty inventories. Preserve
+artifact SHA-256 and index widths. Unrecognized magic may mean encryption and
+is unsupported, not decrypted. Linking names to `GameAssembly` or
+`UnityFramework` code addresses remains a separate native-analysis question.
+See [IL2CPP metadata](https://github.com/morluto/rea/blob/main/docs/il2cpp-metadata.md).
+
+## Rust dependency metadata
+
+Use `inspect_rust_binary` for an explicit local ELF, PE, or Mach-O (universal
+files need `architecture`) when
+the question concerns the cargo-auditable crate list embedded in `.dep-v0`
+(Mach-O: `__DATA,.dep-v0`). No `open_binary`, Rust installation, or target
+execution is required. The CLI equivalent is `inspect-rust-binary PATH`.
+Preserve artifact SHA-256, package order, dependency indices, and the section
+file range. A missing section is null metadata, not proof the binary has no
+dependencies. An omitted package kind stays null. A rustc commit hash is
+inferred from a read-only `/rustc/<40-hex>/` path and does not by itself
+identify the compiler. See [Rust binaries](https://github.com/morluto/rea/blob/main/docs/rust-binaries.md).
+
+## PE Authenticode certificate tables
+
+Use `inspect_pe_certificates` for an explicit local PE32/PE32+ image when the
+question concerns the Authenticode certificate table, WIN_CERTIFICATE types, or
+embedded PKCS#7 signer and X.509 identity fields. No `open_binary`, Windows SDK,
+or target execution is required. The CLI equivalent is
+`inspect-pe-certificates PATH`. Preserve artifact SHA-256, table file ranges,
+payload hashes, original validity-time spellings, and
+`subject_key_identifier_hex` when the signer uses that form. A missing
+`IMAGE_DIRECTORY_ENTRY_SECURITY` directory is observed absence of that table,
+not proof the image is unsigned in every catalog or overlay sense. Signatures,
+chains, timestamps, and Authenticode digest coverage remain unobserved. See
+[PE Authenticode certificate tables](https://github.com/morluto/rea/blob/main/docs/pe-certificates.md).
+
+## PE import and delay-load tables
+
+Use `inspect_pe_imports` for an explicit local PE32/PE32+ image when the
+question concerns imported DLL and function names, ordinals, delay-load
+descriptors, or the pefile import hash. No `open_binary`, Windows SDK, or
+target execution is required. The CLI equivalent is `inspect-pe-imports PATH`.
+Preserve artifact SHA-256, directory file ranges, original DLL spellings, and
+ILT/IAT RVAs. A missing `IMAGE_DIRECTORY_ENTRY_IMPORT` directory is observed
+absence of that table, not proof the image never imported APIs through another
+table or overlay. Bound snapshots, loaded IAT contents, and export-forward
+resolution remain unobserved. See
+[PE import and delay-load tables](https://github.com/morluto/rea/blob/main/docs/pe-imports.md).
+
+## Windows Shell Link shortcuts
+
+Use `inspect_windows_shortcut` for an explicit local `.lnk` file when the
+question concerns the recorded target path, arguments, working directory, icon
+location, volume serial, network share names, ExtraData blocks, or Tracker
+droid identifiers. No `open_binary`, Windows SDK, COM, or target execution is
+required. The CLI equivalent is `inspect-windows-shortcut PATH`. Preserve
+artifact SHA-256, FILETIME raw decimals, IDList type bytes, and ExtraData
+signatures. Unknown IDList items and ExtraData signatures remain raw with
+SHA-256; they are not guessed. Version-1 UUID MAC and timestamp values are
+labelled derived. A missing optional section is observed absence of that
+structure. See
+[Windows Shell Link shortcuts](https://github.com/morluto/rea/blob/main/docs/guides/native/windows-shortcut.md).
+
+## PE TLS callbacks and load configuration
+
+Use `inspect_pe_load_config` for an explicit local PE32/PE32+ image when the
+question concerns code that runs before the entry point through TLS callbacks,
+or the security cookie, SafeSEH, CFG, EH-continuation and XFG declarations in
+the load configuration. No `open_binary`, Windows SDK, or target execution is
+required. The CLI equivalent is `inspect-pe-load-config PATH`. Preserve artifact
+SHA-256, directory and callback-array file ranges, and callback VAs as
+hexadecimal strings. Fields outside the declared `Size` are `null`. Guard-flag
+mitigations carry `evidence_kind: "inferred"`; enforcement remains unobserved.
+See
+[PE TLS callbacks and load configuration](https://github.com/morluto/rea/blob/main/docs/pe-load-config.md).
+
+## Binary hardening
+
+Use `inspect_binary_hardening` for an explicit local ELF, PE, or Mach-O file
+when the question concerns PIE or ASLR, NX, RELRO, stack-protector or FORTIFY
+signs, control-flow flags, or stripped state. No `open_binary`, pwntools, or
+target execution is required. The CLI equivalent is
+`inspect-binary-hardening PATH`. Universal Mach-O files return one image per
+slice. A missing import is `unknown`, not proof the protection was omitted.
+Authenticode and Mach-O code-signature rows report presence, not validity.
+`hardened_runtime` stays unknown and names `inspect_signature`. See
+[binary hardening](https://github.com/morluto/rea/blob/main/docs/binary-hardening.md).
+
+## ELF note tables
+
+Use `inspect_elf_notes` for an explicit local ELF32/ELF64 image when the
+question concerns PT_NOTE/SHT_NOTE owners, GNU build-id, or GNU ABI-tag
+fields. No `open_binary` or target execution is required. The CLI equivalent
+is `inspect-elf-notes PATH`. Preserve artifact SHA-256, table file ranges,
+descriptor hashes, and original owner spellings. A missing table is observed
+absence of those declarations, not proof the image never carried notes in
+another container. See
+[ELF note tables](https://github.com/morluto/rea/blob/main/docs/elf-notes.md).
+
+## ELF and Mach-O initializer functions
+
+Use `inspect_native_initializers` for an explicit local ELF or Mach-O image
+when the question is which functions run before `main` or at exit: ELF
+`DT_PREINIT_ARRAY`, `DT_INIT`, `DT_INIT_ARRAY`, `DT_FINI`, `DT_FINI_ARRAY` and
+legacy `.ctors`/`.dtors`, or Mach-O `__mod_init_func`, `__init_offsets` and
+`__mod_term_func`. Pass `architecture` for a universal Mach-O. The CLI
+equivalent is `inspect-native-initializers PATH`. PIE slots are resolved
+through RELATIVE, RELR and section relocations, and Mach-O slots through
+chained fixups. Keep each entry's `resolution`: `unresolved` means the value was
+not established, and `unresolved_import` names an external symbol. Page with
+`offset` and `limit`. See
+[ELF and Mach-O initializer functions](https://github.com/morluto/rea/blob/main/docs/guides/native/native-initializers.md).
+
+## ELF unwind function ranges
+
+Use `inspect_elf_unwind_ranges` for an explicit local ELF32/ELF64 image when
+the question concerns function starts and lengths that survive `strip`, which
+functions have an LSDA, or which ranges have no exact symbol. No
+`open_binary` or target execution is required. The CLI equivalent is
+`inspect-elf-unwind-ranges PATH`. Preserve artifact SHA-256, FDE start and
+end, containing section, exact symbol name, CIE index, LSDA, and personality.
+Page later ranges with `offset` and `limit`. A missing `.eh_frame` is
+observed absence of that declaration and of `PT_GNU_EH_FRAME`, not proof the
+image has no functions. Header disagreement is a limitation and does not drop
+decoded ranges. `.debug_frame` is reported present or absent and is not
+decoded. CFA instructions are counted and not evaluated. Indirect EH pointers
+are not dereferenced. See
+[ELF unwind function ranges](https://github.com/morluto/rea/blob/main/docs/guides/native/elf-unwind-ranges.md).
+
+## Apple IPS crash reports
+
+Use `inspect_apple_ips` for an explicit local Apple `.ips` file when the
+question concerns crash metadata, exception and termination facts, threads,
+usedImages, lossless registers, or documented crash classes. No `open_binary`,
+Xcode, atos, or target execution is required. The CLI equivalent is
+`inspect-apple-ips PATH`. Preserve artifact SHA-256, `bug_type` as a string,
+uppercase image UUIDs, and integer literals that exceed IEEE-754 range. Only
+`bug_type` `"309"` yields a crash body; other DiagnosticReports types return
+no crash result. Select `join_path` for one local Mach-O or dSYM to match
+report images to slices by UUID and architecture. Preserve the selected
+artifact identity, slice facts and unjoined reasons; a joined static address
+does not resolve function names or dSYM symbols. See
+[Apple IPS crash reports](https://github.com/morluto/rea/blob/main/docs/apple-ips.md).
+
+## Linux kernel modules
+
+Use `inspect_kernel_module` for an explicit local ELF32/ELF64 relocatable
+object when the question concerns `.modinfo` license, vermagic, alias,
+depends, or parm strings, the `__versions` symbol CRCs, or an appended module
+signature trailer. No `open_binary`, Linux host, or `modinfo` program is
+required. The CLI equivalent is `inspect-kernel-module PATH`. Preserve
+artifact SHA-256, key order, duplicate keys, CRC hex, and section file ranges.
+A missing section is observed absence when section names are known, not proof
+the module never carried that metadata. `.gnu.linkonce.this_module` is reported
+as present or absent; `struct module` stays unknown because its layout changes
+with the kernel version. `__ksymtab` records are not decoded. The PKCS#7 range
+and its SHA-256 are observed; certificate, signer, and digest coverage remain
+unvalidated. An absent trailer is not proof the module is unsigned in another
+container. gzip, xz, and zstd whole-file containers are unsupported and name
+`external_decompression` as the next probe: use an external codec utility to
+decompress into a separate file, then call `inspect_kernel_module` on that file. See
+[Linux kernel modules](https://github.com/morluto/rea/blob/main/docs/kernel-modules.md).
+
+## Linux kernel build config
+
+Use `inspect_kernel_build_config` for an explicit local kernel image (`vmlinux`
+ELF, raw arm64 `Image`, x86 `bzImage`/`zImage`, or `configs.ko`) when the
+question is which kernel version it is or how it was configured. No
+`open_binary`, Linux host, or `extract-ikconfig` is required. The CLI
+equivalent is `inspect-kernel-build-config PATH`; pass `name` for one exact
+`CONFIG_*` option and `offset`/`limit` to page the rest. Preserve the banner
+version, builder, and compiler, the config SHA-256 and option count, and each
+entry's value kind (`y`, `m`, `string`, `number`, `not_set`). A config is
+accepted only when `IKCFG_ST` is immediately followed by a valid gzip member
+and `IKCFG_ED`, so an ordinary `IKCFG_ST` string is not a config. Offsets with
+`space: "inflated_payload"` are positions in an inflated bzImage payload, not
+file offsets. An absent config describes only the searched bytes; the kernel
+may have been built without `CONFIG_IKCONFIG` or with it as a module. Truncated
+payloads are incomplete, not absent. A framed blob that does not decode, has a
+bad CRC, or holds non-config text is `malformed`, not absent. xz, lz4, lzma, zstd, bzip2, and lzo
+payloads are unsupported and name `external_decompression`: decompress with an
+external utility, then call `inspect_kernel_build_config` on the result. See
+[Linux kernel build config](https://github.com/morluto/rea/blob/main/docs/guides/native/kernel-build-config.md).
+
+## CPython pyc files
+
+Use `inspect_python_pyc` for an explicit local CPython 3.7–3.13 `.pyc` when the
+question concerns the header magic, PEP 552 source hash or timestamp, nested
+code-object names, or bytecode identity. No `open_binary`, Python installation,
+or target execution is required. The CLI equivalent is `inspect-python-pyc PATH`.
+Preserve artifact SHA-256, the 16-byte header fields, nested code names, and
+bytecode SHA-256. Bytecode is not decompiled. A decoded header does not prove
+the matching `.py` is present. See [CPython pyc inspection](https://github.com/morluto/rea/blob/main/docs/python-pyc.md).
+
+## Erlang BEAM files
+
+Use `inspect_erlang_beam` for an explicit local `FOR1`/`BEAM` file when the
+question concerns IFF chunks, atom names, exports, imports, lambdas, or the
+Code header. No `open_binary`, Erlang installation, or bytecode execution is
+required. The CLI equivalent is `inspect-erlang-beam PATH`. Preserve artifact
+SHA-256, chunk file ranges, 1-based atom indices, and original atom spellings.
+Missing optional tables stay null. For a few facts, use `detail: "summary"` with
+a `beam` view selecting a collection, exact names/indices and fields; read later
+pages from `parent_evidence_id` or the complete Evidence bundle. Same-name
+exports retain every arity and original atom index. CLI summary delivery requires
+`--evidence_output`. Null table counts remain distinct from decoded empty tables. See
+[Erlang BEAM files](https://github.com/morluto/rea/blob/main/docs/erlang-beam.md).
+
+## SQLite snapshots
+
+Use `inspect_sqlite_database` with an explicit local database snapshot for its
+schema and optional rows from one exact ordinary table. It requires Node.js
+24.x >=24.15 or 26+ with the native `DatabaseSync.limits` API. It also consumes the
+existing sibling `-wal`; retain both original digests when citing its result.
+Default output is a compact inventory; a selected table expands only its own
+schema and samples 20 rows. Select `columns` to avoid irrelevant or large cells,
+and request `schema_scope: "database"` only for complete schema details.
+Choose `row_limit` when a finite sample is useful and keep truncation and
+unspecified row order explicit. Virtual tables, shadow tables and views are
+schema facts, not row-read targets. REA inspects private copies and leaves
+the source files untouched; obtain a coherent quiescent database/WAL snapshot
+before inspection. See [SQLite snapshots](https://github.com/morluto/rea/blob/main/docs/guides/data/sqlite-databases.md).
+
+## Recorded Linux crashes
+
+Use `inspect_recorded_crash` with an explicit Linux x86-64 ELF core path to
+inspect historical thread registers, signals, raw notes and original file
+ranges. It is independent of the active disassembler target. Source-bound
+register values and note bytes are observations; missing notes, signal-thread
+association and current executable/library identity remain unknown. PIDs are
+historical metadata and never authorize live attach or control.
+
+The optional `include_debugger_context` facet adds mapping candidates through
+caller-supplied GDB/pwndbg in an owned core-only session. Display names do not
+establish current file identity; zero reported flags leave permissions unknown.
+Requested unavailable context fails with setup guidance. See
+[recorded crashes](https://github.com/morluto/rea/blob/main/docs/guides/native/recorded-crashes.md)
+for exact upstream profiles, bounds and verification coverage.
+
+## Offline ELF layout
+
+Use `inspect_binary_layout` with an explicit local ELF path for file-backed
+sections, symbols, relocations, linkage names, and static mitigation inferences
+without opening a disassembler target. When that complete record is already
+retained, call `inspect_analysis_view` for a summary, mitigations or linkage
+facet, one section or symbol, or a stable page. Do not re-run the decoder to
+read a single object.
+
+## Offline WASM
+
+Use `inspect_wasm_artifact` with caller-supplied WABT for selected local bytes.
+For counts, validation and WAT identity, select `detail: "summary"` on the first
+MCP call. The complete original Evidence stays retained; use
+`normalized_result.parent_evidence_id` for exact exports or text pages. Omitted
+or complete detail keeps the existing complete result.
+For a completed result, use `inspect_analysis_view` with its reported retained
+reference for a summary or `view: {kind: "wasm", facet: "exports", offset: 0, limit: 32}`.
+Facets also include sections, imports, candidates, glue, WAT, headers, details
+and tool profile. Text offsets count UTF-16 code units; follow `next_offset`.
+The complete source remains retained after an oversized MCP reply. No target
+execution or network acquisition is implied, and WAT is decoded representation.
+
+## Offline APK signatures
+
+Use `inspect_apk_signature` for a selected local APK/ZIP/AAB. Default MCP summaries
+retain complete Evidence. Select `view.collection: signers/digests/lineage`,
+with scheme or ID batches, fields and paging; use `inspect_analysis_view` kind
+`apk-signature` for retained reads after deletion. IDs retain artifact/scheme
+scope and signer links; `fields: []` keeps identity links. Empty pages preserve
+partial decoding reasons and unknown authenticity. CLI compact delivery needs
+`--evidence_output` to preserve the complete parent.
+It reads the APK
+Signing Block (v2/v3/v3.1) and v1 `META-INF` signing without executing anything,
+and reports each signer's certificate facts (for v1, the certificate the PKCS#7
+block's SignerInfo names), the declared v3 rotation lineage,
+the declared source-stamp certificate and, for SHA-256/SHA-512 digests, whether
+the declared content digest matches the current bytes. That match is not a
+signature verification: anyone who changes the bytes can recompute the digest.
+`posture` summarises v1-only signing (which cannot tell a v1-only APK from one
+whose v2/v3 block was removed), observed signer agreement across schemes,
+whether the v1 signer equals the block-scheme signer (`false` is notable),
+whether every differing signer appears in a declared (unverified) rotation
+lineage, and the strongest scheme.
+It never claims authenticity: no signature or lineage is cryptographically
+verified.
+
 ## Native targets
 
-For an existing IDA MCP configuration, use `open_binary` with the original
-input binary and `provider_id: "ida"`; do not pass an `.idb` or `.i64` database.
-The legacy attached profile binds the already-open GUI input by SHA-256 and
-leaves its database open. The modern headless profile creates and closes an
-owned private database without saving. Installation and registration reuse
-[mrexodia/ida-pro-mcp](https://github.com/mrexodia/ida-pro-mcp); see the
-[IDA provider guide](https://github.com/morluto/rea/blob/main/docs/providers/ida-provider.md).
-For migration to the separate native IDA MCP plugin, follow the guide's
-[legacy attached migration](https://github.com/morluto/rea/blob/main/docs/providers/ida-provider.md#migrate-a-legacy-attached-registration).
-Replace the upstream registration explicitly; modern mrexodia tools are not the
-native profile. Native attachment preserves the database but does not provide
-direct callers; check session availability and the guide's verification limits.
-Proceed directly to `analyze_function`, pseudocode, or function/string searches
-after opening; IDA does not supply `binary_overview`. Consult session availability
-when composing broader workflows. Modern direct callers and unsupported dossier
-facets remain unknown; live IDA results are not replayed from snapshots.
-
-With `REA_KUNA_COMMAND` configured, `provider_id: "kuna"` answers procedure
-lists, lookups, pseudocode, assembly and `analyze_function` from Kuna's cached
-whole-program export. Callers/callees use a separate body-free graph shared by
-the session; `xrefs` returns exact-address incoming sites and filters forwarding
-aliases. These queries and inventory lookups do not trigger a C export. Indirect
-targets remain unknown and can make relationship coverage partial. Reference
-caches expire on close; a fresh CLI process repeats the native reference walk.
-Typed references, CFG, raw-memory tools and annotations remain unsupported. A cold first
-answer can come from a single-function run (its limitations say so); assembly
-waits for the export. For pseudocode and assembly, `raw_result.source` names the
-cached `.c`/`.asm` file and byte span, which you may search directly instead of
-calling per function. See the
-[Kuna provider guide](https://github.com/morluto/rea/blob/main/docs/kuna-provider.md).
-
 After `open_binary`, use focused search, procedure, or function tools directly.
 Use `binary_overview` when target metadata or inventory context helps answer the
 question. Prefer literal search, names, decompilation, callers, callees, and
@@ -348,8 +632,6 @@ that a class is a root class or has no ivars.
 
 ## Managed PE/CLI
 
-<a id="managed-pe-cli-guide"></a>
-
 Start with `inspect_managed_artifact`. REA's canonical managed inspection is
 execution-free: do not claim it loaded, reflected, executed, or resolved the
 assembly. Keep managed/native boundaries and unavailable reconstruction facts
@@ -408,17 +690,81 @@ and distinguish derived metadata from Ghidra database observations. Recovered
 types and native pseudocode are not original C# source. Windows Ghidra P0 has
 no database mutation authority.
 
-For headerless firmware and address-bearing containers, read
-[raw images](firmware-and-raw-images.md#headerless-raw-images-and-firmware-containers).
+## Packages and extraction
 
-## DOS COM loader interpretation
+Use `inspect_artifact` for application bundles, archives, ZIP/APK/IPA/MSIX/AppX,
+ASAR, or DMG inputs when the artifact graph and findings help answer the
+question. It returns the complete artifact graph inline. Cite graph manifest
+IDs when using them.
+
+For an IPA or a macOS `.app`, ZIP, or DMG, pass the inventory Evidence to
+`project_apple_application_graph`. It reports application roots and nested
+bundles by path convention: app extensions, XPC services, frameworks, login
+items, system and driver extensions, and plug-ins. It also lists privileged
+helpers, launchd plists, symlinks, and each bundle's `info_plist_path` and
+executable candidates. Roles are path conventions, not parsed plists; read the
+listed plists with `inspect_plist`.
+
+After `open_binary` on a `.app` or Mach-O, use `trace_dylib_resolution` to see
+which file each `@rpath`, `@loader_path` and `@executable_path` load reaches for
+every executable in the bundle. It also lists missing or weak loads and earlier
+`@rpath` candidates that are absent. Narrow large bundles with `roots` or
+`architecture`. System paths stay undetermined because the dyld shared cache
+provides them.
+
+Use `extract_artifact` when materialized files from a supported format are needed.
+It extracts the active target and accepts optional `integrity_policy`; see the [MCP contracts](https://github.com/morluto/rea/blob/main/docs/reference/mcp-contracts.md).
+It materializes all regular files into a fresh temporary directory chosen by
+REA. Symlinks and encrypted entries are inventory facts, not extractable files.
+The result reports the materialized directory; callers do not choose the
+destination. No separate permission grant is needed.
+
+For classic iNES cartridges, `open_binary` recognizes the byte signature without
+an executable architecture. `inspect_artifact` retains header facts and each
+PRG/CHR file unit's offset, length and SHA-256; its nested inventory's
+`ines.reconstruction` verifies the actual ordered region bytes against the
+whole-file digest. Extraction includes the original header, optional trainer
+and uninterpreted trailing bytes. These file units are not mapper/CPU bank
+maps. NES 2.0, UNIF, dirty legacy headers and truncated payloads are explicitly
+refused. No emulator or 6502 analysis is performed. See the
+[iNES guide](https://github.com/morluto/rea/blob/main/docs/ines-containers.md).
+
+Native DMG traversal automatically uses a read-only mount on supported macOS
+hosts, with an owned temporary mount directory and cleanup. It needs no approval
+flag. Unsupported hosts or mount failures remain explicit limitations; do not
+claim child inventory when only root identity is available. DMG extraction is
+unavailable through `extract_artifact`. To access DMG files on macOS, create a
+read-only mount with your own tools, select files from that mount for analysis,
+and detach the mount you created when finished. REA's inspection mount is
+temporary, not a retained export of the image's contents.
 
 For headerless DOS COM, explicitly open with `format: "dos-com"` (CLI
 `--target-format dos-com`). Inspect the load image before trusting function
 analysis: the entry is `0x10100`, file offset zero, with imposed real-mode segment
 context. PSP/stack/device state remains unmodeled. See [DOS guide](https://github.com/morluto/rea/blob/main/docs/providers/ghidra-dos.md).
 
-## Function annotations in Ghidra
+Use `inspect_byte_entropy` to triage high- and low-entropy ranges before extraction.
+
+For other headerless images (embedded firmware without ELF/Mach-O/PE/MZ
+headers), open with `format: "raw-binary"`, `architecture` (`x86`, `x86_64`,
+`arm`, `arm64`, `6502`, or `m68k`), and `entry_point` (`0x` hex). File offset 0 is mapped at
+that address. Hopper and IDA refuse this interpretation; Ghidra uses
+`BinaryLoader` with a generic language: big-endian for `m68k`, little-endian
+otherwise. Big-endian ARM, Thumb-only, and Cortex-M variants are unsupported.
+`6502` selects NMOS 6502; the entry and complete file mapping must fit within
+`0x0000..0xffff`. 65C02, 65816, cartridge containers, bank switching, and mapper
+semantics are unsupported. `m68k` requires an even entry and decodes the 68040
+superset of the 68000 family; Amiga/Atari containers and vector-table entry
+discovery are unsupported. Windows Ghidra P0 admits raw images only for `6502`. See [raw binary guide](https://github.com/morluto/rea/blob/main/docs/providers/raw-binary.md).
+
+Intel HEX, S-record, UF2 and DfuSe files are containers, not raw images:
+`inspect_artifact` lists address regions (UF2 per familyID, DfuSe per element
+with its alternate setting), and `extract_artifact` writes each region. Open an
+extracted region as `raw-binary` with its start address as `entry_point`. UF2
+numbering gaps leave coverage unknown; a DfuSe suffix CRC mismatch follows
+`integrity_policy`. See [UF2 and DfuSe](https://github.com/morluto/rea/blob/main/docs/uf2-dfuse.md).
+
+### Function annotations in Ghidra
 
 Fresh-import sessions admit annotations; existing-project reuse is read-only.
 Use `annotate_native_function` for one function name and/or entry comments, with at least one explicit change. Review its annotation readback and refreshed dossier inline. Changes are atomic; empty comments clear them, omitted fields preserve them. Later MCP calls observe edits. Normal close saves them when owned Program checkpoint mode is enabled; ephemeral mode discards them. CLI `annotate-native-function` follows the same setting when closing after its updated result. Select the [checkpoint workflow](#continue-editable-ghidra-analysis-across-processes) before edits that must survive a later process. Original executable bytes are unchanged; edits invalidate immutable snapshots. Windows P0 does not admit database mutations. For naming only, `set_address_name` / `set_addresses_names` also work with Ghidra: they name function entries or data addresses (primary labels) and return compact per-address success booleans instead of a dossier. Ghidra rejects batch addresses that request the same name or a name another selected address holds; `raw_result.failures` gives each failure's reason.
@@ -438,39 +784,160 @@ architecture, provider-owned mapping and selected-image commitments. Equal
 numeric addresses alone cannot establish correspondence; inspect retained
 source Evidence and coverage when the match remains ambiguous.
 
-## Other artifact guides
+## Offline PE resource inventory
 
-Read the owning page for offline formats, containers or firmware. These explicit
-anchors preserve older links without creating duplicate searchable sections.
+Use `inspect_pe_resources` with an absolute `path`, or
+`rea inspect-pe-resources ./selected.exe --json`, for portable PE32/PE32+
+resource inspection without opening a native target. Preserve returned Evidence:
+type/name/language IDs and UTF-16 names, original directory/data/payload ranges,
+code pages, payload hashes and static icon-group references. Same-language icon
+resolution is explicit; missing references and size mismatch remain visible.
+Other payloads are opaque, and this operation does not decode images or simulate
+Windows resource loading. Consult the budgets and supported tree profile in
+[offline binary diagnostics](https://github.com/morluto/rea/blob/main/docs/guides/native/binary-diagnostics.md).
 
-- <a id="go-build-metadata"></a>[Go build metadata](format-inspectors.md#go-build-metadata)
-- <a id="model-files"></a>[Model files](format-inspectors.md#model-files)
-- <a id="il2cpp-metadata"></a>[IL2CPP metadata](format-inspectors.md#il2cpp-metadata)
-- <a id="rust-dependency-metadata"></a>[Rust dependency metadata](format-inspectors.md#rust-dependency-metadata)
-- <a id="pe-authenticode-certificate-tables"></a>[PE Authenticode certificate tables](format-inspectors.md#pe-authenticode-certificate-tables)
-- <a id="pe-import-and-delay-load-tables"></a>[PE import and delay-load tables](format-inspectors.md#pe-import-and-delay-load-tables)
-- <a id="pe-tls-callbacks-and-load-configuration"></a>[PE TLS callbacks and load configuration](format-inspectors.md#pe-tls-callbacks-and-load-configuration)
-- <a id="binary-hardening"></a>[Binary hardening](format-inspectors.md#binary-hardening)
-- <a id="elf-note-tables"></a>[ELF note tables](format-inspectors.md#elf-note-tables)
-- <a id="apple-ips-crash-reports"></a>[Apple IPS crash reports](format-inspectors.md#apple-ips-crash-reports)
-- <a id="linux-kernel-modules"></a>[Linux kernel modules](format-inspectors.md#linux-kernel-modules)
-- <a id="cpython-pyc-files"></a>[CPython pyc files](format-inspectors.md#cpython-pyc-files)
-- <a id="erlang-beam-files"></a>[Erlang BEAM files](format-inspectors.md#erlang-beam-files)
-- <a id="sqlite-snapshots"></a>[SQLite snapshots](format-inspectors.md#sqlite-snapshots)
-- <a id="recorded-linux-crashes"></a>[Recorded Linux crashes](format-inspectors.md#recorded-linux-crashes)
-- <a id="offline-elf-layout"></a>[Offline ELF layout](format-inspectors.md#offline-elf-layout)
-- <a id="offline-wasm"></a>[Offline WASM](format-inspectors.md#offline-wasm)
-- <a id="offline-apk-signatures"></a>[Offline APK signatures](format-inspectors.md#offline-apk-signatures)
-- <a id="packages-and-extraction"></a>[Packages and extraction](packages-and-archives.md#packages-and-extraction)
-- <a id="offline-pe-resource-inventory"></a>[Offline PE resource inventory](format-inspectors.md#offline-pe-resource-inventory)
-- <a id="offline-pe-and-elf-export-tables"></a>[Offline PE and ELF export tables](format-inspectors.md#offline-pe-and-elf-export-tables)
-- <a id="godot-pck"></a>[Godot PCK](packages-and-archives.md#godot-pck)
-- <a id="esp-images"></a>[ESP images](firmware-and-raw-images.md#esp-images)
-- <a id="esp-flash-images"></a>[ESP flash images](firmware-and-raw-images.md#esp-flash-images)
-- <a id="bun-standalone-executables"></a>[Bun standalone executables](packages-and-archives.md#bun-standalone-executables)
-- <a id="release-exposure"></a>[Release exposure](format-inspectors.md#release-exposure)
-- <a id="windows-shell-link-shortcuts"></a>[Windows Shell Link shortcuts](format-inspectors.md#windows-shell-link-shortcuts)
-- <a id="elf-and-mach-o-initializer-functions"></a>[ELF and Mach-O initializer functions](format-inspectors.md#elf-and-mach-o-initializer-functions)
-- <a id="elf-unwind-function-ranges"></a>[ELF unwind function ranges](format-inspectors.md#elf-unwind-function-ranges)
-- <a id="linux-kernel-build-config"></a>[Linux kernel build config](format-inspectors.md#linux-kernel-build-config)
-- <a id="deno-compile-executables"></a>[Deno compile executables](packages-and-archives.md#deno-compile-executables)
+## Offline PE and ELF export tables
+
+Use `inspect_native_exports` with an absolute `path`, or
+`rea inspect-native-exports ./libexample.so --json`, to read a PE export
+directory or ELF dynamic symbol table on any host without opening a native
+target. Ordinal-only, forwarded and hidden-version exports stay distinct;
+partial or unsupported coverage means the inventory is incomplete, not that
+exports are absent. Use `inspect_macho` for Mach-O. See
+[offline binary diagnostics](https://github.com/morluto/rea/blob/main/docs/guides/native/binary-diagnostics.md).
+
+For exact core-module import/export differences, use `compare_wasm_interfaces`
+with two complete inspection Evidence records (or current-session retained
+references) and explicit scope. Preserve unknown types/duplicate names; static
+interface equality is not runtime compatibility. CLI accepts complete inline
+Evidence in `rea compare-wasm-interfaces input.json --json`, or exact references
+with `--evidence-bundle /absolute/path/evidence.json`. Use the complete parent
+ID when the inspection returned a summary.
+
+## Godot PCK
+
+For a standalone PCK, `open_binary` then `inspect_artifact` with
+`detail: "summary"` retains the complete static inventory. Use the returned
+`parent_evidence_id` with `inspect_analysis_view` inventory occurrences filtered
+by exact `logical_paths` or `occurrence_ids`, and join `artifact_id` to nodes
+filtered by `artifact_ids`. Select only needed fields; identity links remain.
+Prefer `inspect-artifact --detail summary --evidence_output PATH` for CLI tasks;
+the same CLI discovers embedded packs via their EOF trailer without opening a
+native analysis provider. Normal native opening remains unchanged.
+
+Distinguish declared MD5 from observed SHA-256 and graph/page completeness from
+`source_coverage`. Encrypted directories and sparse bundles have unknown entry
+coverage; encrypted members have unknown content identity. An empty filtered
+page cannot prove absence in unknown coverage. Retained views are historical,
+not fresh filesystem checks. No decryption, Godot resource decoding, GDScript
+decompilation or target execution is provided. See
+[Godot PCK](https://github.com/morluto/rea/blob/main/docs/godot-pck.md).
+
+UnityFS AssetBundle containers (`.bundle`, `.ab`, YooAsset packages and other
+byte-identified files) list `CAB-*` SerializedFiles and `.resS` members with
+header and node facts; none/LZ4/LZ4HC member blocks stream on demand. Bundles
+declaring asset-bundle encryption keep header facts with unknown directory
+coverage; LZMA member bytes retain named occurrences with unavailable hashes.
+Virtual member offsets are not physical carrier locations. Format 6–8 readers
+cache one decoded block and support retained filtered/paged views. No
+decryption or SerializedFile object decoding is provided. See
+[UnityFS](https://github.com/morluto/rea/blob/main/docs/unityfs.md).
+
+WeChat mini-program packages (`.wxapkg` files with `0xBE`/`0xED` marks) list
+members with declared names and absolute carrier offsets; member bytes are
+stored verbatim. `V1MMWX` PC-client packages report `encrypted` directory
+status with unknown coverage and are refused on extraction; no decryption is
+attempted. See
+[wxapkg](https://github.com/morluto/rea/blob/main/docs/wxapkg.md).
+
+U-Boot legacy uImages use the same inspection/extraction workflows. Prefer a
+retained summary and selected inventory views for header facts or component
+IDs. Header addresses/CPU/OS are declarations; payloads remain as stored.
+Multi-file components retain their aligned physical ranges; the full payload
+is also materialized. Unknown name text retains its original 32 bytes. See
+[U-Boot uImage](https://github.com/morluto/rea/blob/main/docs/uimage.md).
+
+## ESP images
+
+For an ESP32-family app or bootloader image, `open_binary` then
+`inspect_artifact` with `detail: "summary"`. `root_format_metadata` holds the
+observed `header` and `descriptor` (its `kind` is the app or bootloader role)
+and the derived `chip`, `checksum`, `appended_sha256` and
+`undescribed_trailing` ranges. Segment rows (`segment-NN.bin`) carry
+`load_addr`, `classification` (`mapped`, `loaded`, `not-loaded` or `unknown`,
+per ESP-IDF `should_map`/`should_load`; RAM ranges are not checked) and
+`source_location`; select them with `inspect_analysis_view` and
+`logical_paths`. Signature blocks are `unverified`; `public_key_sha256`
+matches eFuse key digests.
+
+Unknown `source_coverage` means truncation or undescribed trailing bytes, such
+as the flash segments of a `--ram-only-header` image; do not infer absence
+from it. Erased flash is described. Integrity mismatches are facts, not
+refusals. ESP8266 images are refused, and flash-encrypted images are not
+recognized. Segments are not disassembled, and RISC-V or Xtensa segments
+cannot be opened as raw-binary. See
+[ESP images](https://github.com/morluto/rea/blob/main/docs/esp-image.md).
+
+## ESP flash images
+
+A whole-flash dump opens as `esp-flash-image`. `root_format_metadata` holds
+`partition_table` (its `method`: `default`, `caller` or a derived
+`scan`), `bootloader` and `apps` ESP image facts, the `otadata` records,
+`partitions_outside_dump` and `boot_selection`. Every table has a
+`boot_selection`: its `basis` names the bootloader's branch (`no-otadata`
+without otadata, `none` when the bootloader rejects the table, `unknown`
+when a record is missing), and it lists its `assumptions`;
+`running_partition` is always `unknown`. Each image names its entry
+`path`. With `detail: "summary"` the root kind is `esp-flash-image-summary`
+and each image is a compact `summary` (chip, entry point, segment count,
+classified load ranges, signature block count, descriptor identity,
+integrity); complete detail keeps every segment. Partition rows
+(`partitions/<label>.bin`) carry the declared `type`, `subtype`, names,
+`flash_address`, `size` and flags; `encrypted` is a storage flag, not
+proof of ciphertext. `written_offset`/`written_end` bound the bytes that
+are not erased: an app slot whose facts say `erased` can still hold an older
+image, and a limitation says so. For a dump read from another flash address, call
+`open_binary` with `format: "esp-flash-image"` and `flash_offset`; an
+`esp-image` whose `flash_dump` is set names that offset in a next probe; a
+relocated table can be named with `partition_table_offset`. No table found
+means unknown, often flash encryption. Extraction writes `nvs_keys` as
+ordinary data. See
+[ESP flash images](https://github.com/morluto/rea/blob/main/docs/esp-flash-image.md).
+
+## Bun standalone executables
+
+A `bun build --compile` executable stores its module graph in Mach-O
+`__BUN,__bun`, a PE or ELF `.bun` section, or an appended trailer.
+`open_binary` then `inspect_artifact` reports root format `bun-standalone` with
+the carrier, structural layout, module count, entry point, raw flags and
+compile exec-argv; each module keeps its `/$bunfs/...` source name, loader,
+encoding, sizes and SHA-256, with sourcemap, bytecode and module-info regions as
+separate children. Use summary root metadata `entry_point_logical_path` as the
+exact inventory `logical_paths` selector, then read the selected content node by
+its artifact ID; the reader owns virtual-name normalization and indexed fallback. `extract-artifact` writes stored contents through the safe
+output tree; route extracted JavaScript to `analyze_javascript_application`.
+Layout labels describe record shapes, not a detected compiler version; unknown
+or ambiguous layouts are unsupported. Bytecode is retained, not decoded, and the
+executable is never run. See
+[Bun standalone executables](https://github.com/morluto/rea/blob/main/docs/bun-standalone.md).
+
+## Deno compile executables
+
+A `deno compile` executable stores its application in a Mach-O `__SUI,d3n0l4nd`
+section, an Intel Mach-O `<~sui-data~>` trailer, an ELF `SUI` note of type
+`0x53554901`, or a PE `RT_RCDATA` resource named `D3N0L4ND`.
+`open_binary` then `inspect_artifact` reports root format `deno-standalone`
+with the carrier, entry point, recorded options, and one occurrence per virtual
+file, including size and SHA-256. Transpiled output, CJS export analysis, and
+source maps are separate occurrences with the file's `index` and their own
+`role`. They normally use a `.deno-*` suffix on the file path and move to an
+indexed path on a collision, so use the reported `logical_path`. Symlink targets
+stay recorded path parts. A virtual filesystem that cannot be listed leaves
+coverage unknown.
+Environment-file values and CA certificate bytes are retained in local Evidence
+with their counts and byte lengths. The npm snapshot length is reported and not
+decoded. `extract-artifact` writes
+stored files through the safe output tree; route extracted JavaScript to
+`analyze_javascript_application`. Other layouts are unsupported, and the
+executable is never run. See
+[Deno compile executables](https://github.com/morluto/rea/blob/main/docs/deno-standalone.md).

.cursor-plugin/skills/reverse-engineer-anything/references/packages-and-archives.md

@@ -1,146 +0,0 @@
-# Packages and archives
-
-Read only the sections needed for the selected task.
-
-- [Packages and extraction](#packages-and-extraction)
-- [Godot PCK](#godot-pck)
-- [Bun standalone executables](#bun-standalone-executables)
-- [Deno compile executables](#deno-compile-executables)
-
-## Packages and extraction
-
-For PyInstaller executables, the CArchive is the application layer; the native
-bootloader is only its prefix. Prefer summary delivery for large inventories,
-then retained inventory views with selected member fields/pages. Runtime options
-are metadata, not files. Compact delivery still inventories and hashes members.
-A small complete inventory can cost less than extra calls and schemas. See the
-[Python packaged applications guide](https://github.com/morluto/rea/blob/main/docs/python-packaged-applications.md).
-
-Use `inspect_artifact` for application bundles, archives, ZIP/APK/IPA/MSIX/AppX,
-ASAR, or DMG inputs when the artifact graph and findings help answer the
-question. It returns the complete artifact graph inline. Cite graph manifest
-IDs when using them.
-
-For an IPA or a macOS `.app`, ZIP, or DMG, pass the inventory Evidence to
-`project_apple_application_graph`. It reports application roots and nested
-bundles by path convention: app extensions, XPC services, frameworks, login
-items, system and driver extensions, and plug-ins. It also lists privileged
-helpers, launchd plists, symlinks, and each bundle's `info_plist_path` and
-executable candidates. Roles are path conventions, not parsed plists; read the
-listed plists with `inspect_plist`.
-
-After `open_binary` on a `.app` or Mach-O, use `trace_dylib_resolution` to see
-which file each `@rpath`, `@loader_path` and `@executable_path` load reaches for
-every executable in the bundle. It also lists missing or weak loads and earlier
-`@rpath` candidates that are absent. Narrow large bundles with `roots` or
-`architecture`. System paths stay undetermined because the dyld shared cache
-provides them.
-
-Use `extract_artifact` when materialized files from a supported format are needed.
-It extracts the active target and accepts optional `integrity_policy`; see the [MCP contracts](https://github.com/morluto/rea/blob/main/docs/reference/mcp-contracts.md).
-It materializes all regular files into a fresh temporary directory chosen by
-REA. Symlinks and encrypted entries are inventory facts, not extractable files.
-The result reports the materialized directory; callers do not choose the
-destination. No separate permission grant is needed.
-
-For classic iNES cartridges, `open_binary` recognizes the byte signature without
-an executable architecture. `inspect_artifact` retains header facts and each
-PRG/CHR file unit's offset, length and SHA-256; its nested inventory's
-`ines.reconstruction` verifies the actual ordered region bytes against the
-whole-file digest. Extraction includes the original header, optional trainer
-and uninterpreted trailing bytes. These file units are not mapper/CPU bank
-maps. NES 2.0, UNIF, dirty legacy headers and truncated payloads are explicitly
-refused. No emulator or 6502 analysis is performed. See the
-[iNES guide](https://github.com/morluto/rea/blob/main/docs/ines-containers.md).
-
-Native DMG traversal automatically uses a read-only mount on supported macOS
-hosts, with an owned temporary mount directory and cleanup. It needs no approval
-flag. Unsupported hosts or mount failures remain explicit limitations; do not
-claim child inventory when only root identity is available. DMG extraction is
-unavailable through `extract_artifact`. To access DMG files on macOS, create a
-read-only mount with your own tools, select files from that mount for analysis,
-and detach the mount you created when finished. REA's inspection mount is
-temporary, not a retained export of the image's contents.
-
-Use `inspect_byte_entropy` to triage high- and low-entropy ranges before extraction.
-
-## Godot PCK
-
-For a standalone PCK, `open_binary` then `inspect_artifact` with
-`detail: "summary"` retains the complete static inventory. Use the returned
-`parent_evidence_id` with `inspect_analysis_view` inventory occurrences filtered
-by exact `logical_paths` or `occurrence_ids`, and join `artifact_id` to nodes
-filtered by `artifact_ids`. Select only needed fields; identity links remain.
-Prefer `inspect-artifact --detail summary --evidence_output PATH` for CLI tasks;
-the same CLI discovers embedded packs via their EOF trailer without opening a
-native analysis provider. Normal native opening remains unchanged.
-
-Distinguish declared MD5 from observed SHA-256 and graph/page completeness from
-`source_coverage`. Encrypted directories and sparse bundles have unknown entry
-coverage; encrypted members have unknown content identity. An empty filtered
-page cannot prove absence in unknown coverage. Retained views are historical,
-not fresh filesystem checks. No decryption, Godot resource decoding, GDScript
-decompilation or target execution is provided. See
-[Godot PCK](https://github.com/morluto/rea/blob/main/docs/godot-pck.md).
-
-UnityFS AssetBundle containers (`.bundle`, `.ab`, YooAsset packages and other
-byte-identified files) list `CAB-*` SerializedFiles and `.resS` members with
-header and node facts; none/LZ4/LZ4HC member blocks stream on demand. Bundles
-declaring asset-bundle encryption keep header facts with unknown directory
-coverage; LZMA member bytes retain named occurrences with unavailable hashes.
-Virtual member offsets are not physical carrier locations. Format 6–8 readers
-cache one decoded block and support retained filtered/paged views. No
-decryption or SerializedFile object decoding is provided. See
-[UnityFS](https://github.com/morluto/rea/blob/main/docs/unityfs.md).
-
-WeChat mini-program packages (`.wxapkg` files with `0xBE`/`0xED` marks) list
-members with declared names and absolute carrier offsets; member bytes are
-stored verbatim. `V1MMWX` PC-client packages report `encrypted` directory
-status with unknown coverage and are refused on extraction; no decryption is
-attempted. See
-[wxapkg](https://github.com/morluto/rea/blob/main/docs/wxapkg.md).
-
-U-Boot legacy uImages use the same inspection/extraction workflows. Prefer a
-retained summary and selected inventory views for header facts or component
-IDs. Header addresses/CPU/OS are declarations; payloads remain as stored.
-Multi-file components retain their aligned physical ranges; the full payload
-is also materialized. Unknown name text retains its original 32 bytes. See
-[U-Boot uImage](https://github.com/morluto/rea/blob/main/docs/uimage.md).
-
-## Bun standalone executables
-
-A `bun build --compile` executable stores its module graph in Mach-O
-`__BUN,__bun`, a PE or ELF `.bun` section, or an appended trailer.
-`open_binary` then `inspect_artifact` reports root format `bun-standalone` with
-the carrier, structural layout, module count, entry point, raw flags and
-compile exec-argv; each module keeps its `/$bunfs/...` source name, loader,
-encoding, sizes and SHA-256, with sourcemap, bytecode and module-info regions as
-separate children. Use summary root metadata `entry_point_logical_path` as the
-exact inventory `logical_paths` selector, then read the selected content node by
-its artifact ID; the reader owns virtual-name normalization and indexed fallback. `extract-artifact` writes stored contents through the safe
-output tree; route extracted JavaScript to `analyze_javascript_application`.
-Layout labels describe record shapes, not a detected compiler version; unknown
-or ambiguous layouts are unsupported. Bytecode is retained, not decoded, and the
-executable is never run. See
-[Bun standalone executables](https://github.com/morluto/rea/blob/main/docs/bun-standalone.md).
-
-## Deno compile executables
-
-A `deno compile` executable stores its application in a Mach-O `__SUI,d3n0l4nd`
-section, an Intel Mach-O `<~sui-data~>` trailer, an ELF `SUI` note of type
-`0x53554901`, or a PE `RT_RCDATA` resource named `D3N0L4ND`.
-`open_binary` then `inspect_artifact` reports root format `deno-standalone`
-with the carrier, entry point, recorded options, and one occurrence per virtual
-file, including size and SHA-256. Transpiled output, CJS export analysis, and
-source maps are separate occurrences with the file's `index` and their own
-`role`. They normally use a `.deno-*` suffix on the file path and move to an
-indexed path on a collision, so use the reported `logical_path`. Symlink targets
-stay recorded path parts. A virtual filesystem that cannot be listed leaves
-coverage unknown.
-Environment-file values and CA certificate bytes are retained in local Evidence
-with their counts and byte lengths. The npm snapshot length is reported and not
-decoded. `extract-artifact` writes
-stored files through the safe output tree; route extracted JavaScript to
-`analyze_javascript_application`. Other layouts are unsupported, and the
-executable is never run. See
-[Deno compile executables](https://github.com/morluto/rea/blob/main/docs/deno-standalone.md).

.cursor-plugin/skills/reverse-engineer-anything/references/recorded-captures.md

@@ -1,58 +0,0 @@
-# Recorded captures and decoder discovery
-
-## Historical web captures
-
-Retained HAR/SAZ/native mitmproxy/aioquic qlog capture: `inspect_web_network_capture`. Select
-original record ordinals when useful; inspect producer fields and byte/number
-sidecars without fetching recorded URLs or claiming live attribution. For a decoder heap refusal, follow the reported `capture.decoder_old_generation_mib`
-setting and bounds with `update_analysis_settings` (default 192 MiB, floor 8,
-maximum 16384); the launch seed
-`REA_CAPTURE_DECODER_OLD_GENERATION_MIB` affects future REA processes. Host memory
-and `NODE_OPTIONS` do not raise that decoder allowance. See
-[historical captures](https://github.com/morluto/rea/blob/main/docs/guides/web/web-network-captures.md) for the exact upstream
-profile and credential exclusions. Qlog support starts with aioquic 1.5.0
-JSON 0.3; stream coverage and endpoint loss declarations do not recover bodies
-or establish network loss. Prefer detail summary, then kind qlog trace/stream
-pages; select one trace, stream ID and direction for contribution/range pages.
-Facts retain all-recorded-events-before-selection scope. Other qlog layouts
-and PCAP correlation are unsupported.
-
-## Packet fields and packet captures
-
-Packet-field discovery: `inspect_packet_fields` searches installed protocols
-and field names/types before choosing fields for `inspect_packet_capture`.
-Query/page the relevant records; registrations do not establish observed traffic.
-
-## Packet capture inspection
-
-Recorded packet captures: `inspect_packet_capture` with selected TShark fields,
-display filters and decode-as rules. Requires Linux, util-linux prlimit and
-TShark 4.x on an absolute `PATH` entry or via the optional absolute
-`REA_TSHARK_COMMAND` override; see
-[capture guide](https://github.com/morluto/rea/blob/main/docs/guides/data/packet-captures.md).
-Packet limits count input packets before filtering. Output defaults to 50 matching
-frames; use `match_offset` / `match_limit` for later pages, or explicit null
-limit for complete matching output. Preserve the source digest across pages.
-This does not perform live capture.
-For address/DNS/SNI/protocol summaries use `inspect_pcap_capture` with MCP
-detail summary, then retained kind pcap pages. Requires Wireshark 4.x tshark
-and capinfos, with Linux prlimit or Windows Job Objects; see
-[packet summaries](https://github.com/morluto/rea/blob/main/docs/guides/data/pcap-analysis.md).
-Select an exact DNS question for address pages; page exhaustion does not prove
-network completeness. This reads recordings and does not acquire traffic.
-
-## Signal decoder discovery
-
-Decoder discovery: `inspect_signal_decoders` searches installed decoder IDs
-or selected decoder help, including channels and options. Query and page the
-relevant details before choosing settings; this does not scan hardware.
-
-## Recorded logic and signals
-
-Recorded logic/signal traces: `decode_signal_recording` with explicit decoder,
-channel and protocol settings. On Linux with util-linux prlimit, supply sigrok-cli 0.7.2 through
-`REA_SIGROK_COMMAND`; see [recording guide](https://github.com/morluto/rea/blob/main/docs/guides/signals/signal-recordings.md).
-Output defaults to 50 literal lines; select annotation classes and page with
-`line_offset` / `line_limit`, or explicit null for all remaining output.
-Window counts are display lines, not annotations or samples. Preserve source digests across pages.
-This reads existing files; it does not acquire hardware signals.

.cursor-plugin/skills/reverse-engineer-anything/references/runtime-observation.md

@@ -1,12 +1,5 @@
 # Runtime observation
 
-Read only the sections needed for the selected task.
-
-- [Native runtime questions](#native-runtime-questions)
-- [Browser and Electron](#browser-and-electron)
-- [Who performs the action](#who-performs-the-action)
-- [Fiddler Everywhere traffic debugging](#fiddler-everywhere-traffic-debugging)
-
 ## Native runtime questions
 
 Start from the requested observation. State the selected target, input, expected
@@ -61,18 +54,6 @@ and [native calls/UI](https://github.com/morluto/rea/blob/main/docs/guides/nativ
 
 ## Browser and Electron
 
-Passive traffic capture of a SPA: the default scope is the page's own origin,
-and data APIs usually live on another origin. Discover API origins from
-absolute URLs in loaded bundles or script text via JavaScript analysis tools
-(another passive capture without those origins still scopes to the page
-origin, and `completeness.excluded` counts lack URLs). Pass both the page
-origin and those API origins in `allowed_origins` — the list replaces the
-default scope — and read `completeness.excluded` before reporting that an
-API is absent. For long captures with body shapes, use
-`export_evidence_bundle` instead of expecting an inline result.
-
-User-owned Electron runtime already open: `list_electron_targets`. Before an electron-main Inspector attempt, check packaged fuses with `inspect_electron_fuses`; disabled `EnableNodeCliInspectArguments` means the app ignores `--inspect`.
-
 Supply the existing browser or Electron instance's literal-port loopback CDP
 endpoint in the request. No separate permission configuration or per-call
 approval flag is needed. Browser discovery lists eligible HTTP(S) pages;

.github/workflows/ci.yml

@@ -755,7 +755,9 @@ jobs:
     needs: [changes, static, build]
     if: needs.changes.outputs.apple == 'true'
     runs-on: macos-15
-    timeout-minutes: 15
+    # Passing runs take 10-15 minutes on macOS runners of varying speed, so a
+    # 15-minute budget cancelled about a third of runs. Keep a hang bounded.
+    timeout-minutes: 25
     steps:
       - uses: actions/checkout@11d5960a326750d5838078e36cf38b85af677262 # v4.4.0
       - uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020 # v4.4.0

bridge/android/ReaJadxBridge.java

@@ -464,7 +464,9 @@ class ReaJadxBridge {
       if (element.getTagName().equals("uses-sdk")) {
         minSdk = element.getAttributeNS(android, "minSdkVersion");
         targetSdk = element.getAttributeNS(android, "targetSdkVersion");
-      } else if (element.getTagName().equals("uses-permission")) {
+      } else if (element.getTagName().equals("uses-permission")
+          || element.getTagName().equals("uses-permission-sdk-23")
+          || element.getTagName().equals("uses-permission-sdk-m")) {
         String permission = element.getAttributeNS(android, "name");
         if (!permission.isEmpty()) permissions.add(permission);
       }

bridge/cutter/rea_bridge.py

@@ -958,15 +958,6 @@ class _BridgeServer(threading.Thread):
             proof.update(fragment.encode("utf-8"))
         return proof.hexdigest()
 
-    def _encode_authenticated_response(
-        self, value: dict[str, Any], client_nonce: str, server_nonce: str
-    ) -> bytes | None:
-        """Return one exact frame. The socket path writes the same pieces."""
-        proof = self._authenticated_proof(value, client_nonce, server_nonce)
-        if proof is None:
-            return None
-        return b"".join(self._iter_authenticated_frame(value, proof))
-
     def _iter_authenticated_frame(self, value: dict[str, Any], proof: str):
         yield _AUTHENTICATED_PREFIX
         for fragment in _response_fragments(value):

bridge/hopper_bridge.py

@@ -154,13 +154,6 @@ class HopperApiFacade:
             )
 
 
-def _configure_hopper_api(document_provider):
-    """Inject Hopper's document provider or a standalone test facade."""
-    global _hopper_api, _rea_session_document
-    _hopper_api = HopperApiFacade(document_provider)
-    _rea_session_document = None
-
-
 def _api():
     """Resolve Hopper globals lazily so importing this module needs no Hopper."""
     global _hopper_api

docs/development/testing.md

@@ -238,7 +238,9 @@ broad changes or diagnosing CI, rather than after every edit. Package/install
 changes additionally need package verification; provider changes need actual
 provider evidence.
 
-Vitest projects use up to two workers, bounded by host parallelism.
+Vitest projects use up to two workers, bounded by host parallelism. GitHub
+Actions jobs own their runner and use up to four; see
+[test deadlines](#developer-commands) for how deadlines scale there.
 `process-boundary` runs later with serial files because process-tree sampling
 shares host resources. Acceptance and process-global cases use isolated forks
 without serial scheduling. Pure domain/service, composition, conformance,
@@ -283,7 +285,7 @@ coverage. Ordinary CI also exhausted the former 384 MiB budget after earlier
 files loaded the shared source/contract graph, while later command registration
 and schema/help rendering still needed transient allocation space. This is a
 measured capacity control, not a suite-wide speedup claim. The worker count
-remains bounded at two, and spawned REA processes retain their own heap settings.
+remains bounded, and spawned REA processes retain their own heap settings.
 Coverage enablement still controls instrumentation, test deadlines and coverage
 thresholds; explicitly disabling it or selecting only a provider does not enable
 instrumentation.
@@ -299,6 +301,21 @@ work from fixture generation, schema compilation, subprocess startup and cleanup
 Choose elapsed, CPU and process-family memory limits before the run; a worker
 count or test timeout alone does not enforce those limits.
 
+A test deadline turns a hang into a named failure; it is not a performance
+assertion. `vitest.config.ts` applies the deadlines in
+`tests/support/testDeadlines.ts` per project: 30 seconds for source projects
+and 120 seconds for built projects, with hooks sharing their project's
+deadline. Coverage doubles every deadline, and so does hosted CI, where
+CPU-bound cases share every core with other workers and their subprocesses.
+Do not pass a per-test timeout to fit a slow host. A child run through a
+test process registry (`processTest`, `cli.run`, or `createTestProcessRegistry`)
+gets the running test's remaining deadline less a reporting reserve, so a hung
+child fails with its captured output. Pass `timeoutMs` only when the child's deadline is itself under test. Tag a bounded workload that needs longer with
+`{ tags: ["long-running"] }`, or a whole file with
+`// @module-tag long-running`. Assert elapsed time explicitly when speed is the
+behavior under test. Derive an explicit deadline only when a test must outlive
+an operation deadline whose diagnostics it checks.
+
 Keep capacity fixtures intact when diagnosing a test failure. For byte equality
 of large buffers, use `Buffer.equals` or `Buffer.compare`; a generic structural
 diff can allocate far more than the production operation.
@@ -1077,32 +1094,12 @@ existing four-minute timing collection plus the bounded diagnostic stages.
 
 Fast built shards target under ten minutes of uncontended shard-job execution.
 That figure is the `test-shard` job. The job waits for classification and the
-build, and the portable-test merge runs after the slowest shard. On 2026-10-20,
-run `39451548945` measured those surrounding jobs with runner queue excluded:
-classification 15 seconds, build 49 seconds, and the portable-test merge 29
-seconds. Shard 1/5 of that run spent 22 seconds before built Vitest and 3
-seconds after it, including `npm ci` at 9 seconds. `node_modules` is about 674
-MiB, so publishing it as a shared artifact would transfer more data than that
-install saves. The npm download cache stays the setup cache.
-
-File timings from that run put the built suite at 7944 seconds of module time.
-The sequencer keeps each file's full duration on one worker. Extra workers
-overlap other files in the same sequence group, and `process-boundary` runs
-after that group. Scheduling the unsplit inventory onto five shards peaks at
-820 seconds of tests. That run's slowest built shard job lasted 925 seconds
-(shard 4/5), and shard 1/5 spent 874 seconds in built Vitest, so hosted job
-time sits above the file schedule. Six shards of the same inventory peak at
-684 seconds of tests. Eight shards, after the 578-second loader file is split
-by case count (54 cases stay in that file and 46 move into each of three
-files), peak at 513 seconds of tests. With the measured 25-second shard setup,
-that shard job is about 538 seconds. Adding the measured classification, build,
-and merge makes an uncontended portable critical path of about 631 seconds,
-still before the gap between this file schedule and hosted Vitest. Runner queue
-is separate and grows when more shard jobs compete for runners.
+build, and the portable-test merge runs after the slowest shard. Runner queue is
+separate and grows when more shard jobs compete for runners. The npm download
+cache stays the setup cache rather than a shared `node_modules` artifact.
 `process-boundary` stays serial because process-tree sampling shares the host.
-Raising the worker cap from two to four on this four-core host cut a six-file
-acceptance sample from 96–101 seconds to 71 seconds and did not remove the
-single-file floor, so the cap stays two. Full qualification keeps six covered
+Hosted CI jobs use four workers; other hosts stay at two
+because they are shared with other checks. Full qualification keeps six covered
 shards; its longer instrumentation is outside this fast budget.
 
 The separate `CI timings` workflow batches recent completed CI and Release runs

docs/guides/apple/apple-application-analysis.md

@@ -215,7 +215,10 @@ Resolution follows dyld:
   image that loaded it, up to the process root.
 - Dependencies load dependents-first in load-command order, and each image
   loads once per process. A later @rpath request whose install name matches an
-  already loaded image reuses it.
+  already loaded image reuses it. Loading an image and expanding its dependencies
+  are separate: an already loaded sibling can first be expanded through another
+  dependency, inheriting that traversal's rpath stack. Cycles do not expand an
+  image twice, and lazy edges do not establish a launch-time traversal chain.
 - Symlinks such as `Versions/Current` are followed inside the analyzed root.
   An escape through `..` or an absolute link is `escapes-target`.
 - Dyld path components are POSIX names. On Windows hosts, link text uses `\`
@@ -252,8 +255,12 @@ one of:
 
 A `conditional` image is still traced, because it is an in-root candidate
 that may load. Every edge below it has `loader_conditional: true`, and a
-request that reuses it by install name is `conditional` too. An in-root dylib
-with no platform load command remains unknown unless the root is an executable
+request that reuses it by install name is `conditional` too. A definitely loaded
+image can also first expand through a conditional chain: its dependency edges
+then have `loader_conditional: true`, while later reuse of the image retains its
+definite first-load result.
+
+An in-root dylib with no platform load command remains unknown unless the root is an executable
 with observed macOS platform and the candidate's CPU type is x86_64, i386, or arm64. In that macOS context,
 [Apple's dyld](https://github.com/apple-oss-distributions/dyld/blob/fd8d0c4d52320ebf64db34f3cb280310d905c5ae/mach_o/UnsafeHeader.cpp#L1137-L1145) supplies an implicit macOS platform to legacy dylibs on those CPU
 types. The candidate records `platform_basis: "macos-dyld-implicit"`; the

docs/reference/mcp-contracts.md

@@ -222,7 +222,7 @@ cannot answer, such as an address without one exact file offset, omits this
 hint; follow its `remediation.action`. `recovery.kind=inspect_timeout_state` names
 the operation and the REA server/provider deadline. Inspect partial Evidence
 and reported cleanup state, then follow the returned remediation. Change a
-request field or REA_* setting only when the returned remediation explicitly
+request field or REA\_\* setting only when the returned remediation explicitly
 recommends that change. Restart REA when that text says the value is read at
 startup. A mention that a control does not raise the deadline is not a
 recommendation to change it. When it names no setting, reduce supported work;
@@ -1065,6 +1065,17 @@ from an older request does not override a newer outcome. Request deadlines and
 cancellations neither mark nor clear availability. A new provider client starts
 without these outcomes.
 
+A `provider_missing` group may also list `alternatives`: advertised tools the
+current provider makes available for part of that operation's question. When
+the provider does not declare `next_address` or `prev_address`, they list
+whichever of `inspect_native_instruction`, `read_function_instructions` and
+`procedure_assembly` qualify. These answer instruction-level questions only:
+none navigates arbitrary analyzed objects such as data or addresses outside a
+procedure, and neither listing order nor address plus length proves executed
+control flow. Calling the missing operation returns `capability_unavailable`
+with the same tools in `details.alternatives` and a remediation stating each
+one's scope. Without a qualifying alternative, both keep their generic advice.
+
 Standalone command operations (such as `inspect_with_objdump` and
 `execute_rizin_command`) similarly reflect observed executable launch failures:
 an unprobed executable is reported available; once a launch failure is observed,
@@ -1911,13 +1922,10 @@ support for omitted historical values.
 known and differ. If only one value is known, storage comparison is unknown;
 when both are omitted, storage is not compared.
 
-<!-- BEGIN:compare-artifacts:compared-format-metadata-kinds -->
-
-`compare_artifacts` also compares the declarations in the `format_metadata`
-of UnityFS members (`unityfs-entry`), PyInstaller entries
+`compare_artifacts` also compares the declarations in the `format_metadata` of <!-- BEGIN:compare-artifacts:compared-format-metadata-kinds -->UnityFS members (`unityfs-entry`), PyInstaller entries
 (`pyinstaller-entry`), Node SEA members (`node-sea-entry`),
 ESP image segments (`esp-image-segment`), DfuSe elements (`dfuse-element`),
-ESP flash partitions (`esp-flash-partition`) and FIT images (`fit-image`);
+ESP flash partitions (`esp-flash-partition`) and FIT images (`fit-image`)<!-- END:compare-artifacts:compared-format-metadata-kinds -->;
 each format's guide names the compared fields. A changed declaration is a
 metadata change, including when content is unavailable. A declaration that
 only one side records, or that the other side records as another kind, is
@@ -1928,8 +1936,6 @@ container or from caller options that identify the target, or repeat
 occurrence fields, except Godot PCK source paths, which
 [Godot PCK archives](../godot-pck.md) explains.
 
-<!-- END:compare-artifacts:compared-format-metadata-kinds -->
-
 Every occurrence records `entry_kind`, `executable` and `encrypted`. A
 different entry kind or `encrypted` value is a `metadata` difference, and a
 different `executable` value is an `executable` difference. Occurrence and node

docs/reverse-engineering-tools.md

@@ -96,6 +96,10 @@ Persistent command Evidence retains the caller-selected and resolved startup pat
 
 Session status retains up to 64 recent protocol frames. Each history entry keeps a complete-character UTF-8 prefix of at most 16 KiB, plus a truncation marker when needed; `recent_output_truncated` reports omitted history. This history budget also accounts for replacement characters from invalid UTF-8. It does not shorten command Evidence, whose output budget defaults to 16 MiB. Rizin debug startup waits 30000 ms with no runtime override. Each command frame defaults to 120000 ms; its deadline, command-output budget, session capacity and pending-command queue use the [shared debugger launch settings](#debugger-launch-settings). The 16777216-byte frame cap and 4096-frame command cap remain fixed. Evidence names the cap that stopped the command. The one-shot `command_timeout_ms` setting does not change a debug session.
 
+### Stopped debugger sessions
+
+A Rizin or GDB session stops accepting commands when REA stops it, for example after an output or frame limit, a command deadline, cancellation or a protocol failure, or when the debugger process exits. REA keeps the stopped session and the first reason it observed until `close_rizin_debug_session` or `close_gdb_session` releases it, including after a retried cleanup succeeds. `rizin_debug_session_status` and `gdb_session_status` then report `state: "closed"` with that reason in `closed_reason`. A later `rizin_debug_command` or `gdb_console` on the session fails with `capability_unavailable`; `details.reason` carries the same reason, and the remediation is to close the session and start a new one. A `session_id` that this server never returned, or that close already released, fails with `invalid_request` instead.
+
 ### Debugger launch settings
 
 GDB and Rizin share these settings in the environment that launches REA:
@@ -136,6 +140,8 @@ rea debug-with-rizin --inferior-tty /dev/pts/7 ./program dr
 rea execute-rizin-command ./program iI
 ```
 
+Rizin exits with status 0 even when its shell refuses a statement, so the exit code does not show whether the command ran. Rizin 0.6 through 0.9.1 log one of four stderr lines for such a refusal: `ERROR: Command '…' does not exist.`, ``ERROR: Wrong number of arguments passed to `…`…``, `ERROR: core: Error while executing command: …` (an invalid statement), or `ERROR: core: Error while parsing command: …`. Rizin then skips the refused statement and every later one in the command. When the completed process's stderr contains one of these lines, REA reports `completion_status: "rejected"` and adds a limitation naming the refusal. stdout then holds only the output of earlier statements, so empty stdout is not an empty result. Other `ERROR:` lines, such as `ERROR: Cannot find function at …`, come from commands that ran; REA keeps them in stderr and reports `complete`. A command that lowers `log.level` or redirects stderr hides refusals from REA.
+
 Captured stdout and stderr use the same `max_output_bytes` (CLI: `--max-output-bytes`) admission and recovery described above, default 16 MiB. One-shot commands independently use `command_timeout_ms`. Raising output admission cannot recover an earlier capture; command effects and cleanup must be checked before another invocation.
 
 ### Decoder and map inputs

docs/specs/desktop-app-environments/file-readers/plan.md

@@ -10,8 +10,8 @@
 4. Verifier: `appTaskPython.ts` renders `READER_RUNNER`, `READERS`,
    `_reader_views` and `evaluate_with_readers` only for a task with readers,
    so a task without renders as before.
-5. Application: `readTaskReaderViews` and `readerViewsOfBytes` in
-   `DesktopTaskReaders.ts`; `runReaderHere` in `src/desktop/TaskReaderProcess.ts`;
+5. Application: `readerViewsOfBytes` in `DesktopTaskReaders.ts`;
+   `runReaderHere` in `src/desktop/TaskReaderProcess.ts`;
    the witness runs readers through `AgentEnvComputer.exec`; the export runs
    them on `shippedSeedBytes` and the end's retained content.
 6. Docs: the tasks reference, the witness page, the build skill's recipes for

docs/testing-lanes.md

@@ -842,28 +842,14 @@ branches are not producers. A tier with no qualifying artifact leaves history
 unset, and Vitest uses its own assignment. A later schedule that reuses
 qualification without uploading timings leaves the previous artifact in place.
 
-On 2026-10-21 the newest full history was run
-`39545047294`, recorded before compiled-CLI session reuse. Fast run
-`39552964873` packed its eight built shards from that history, and those shard
-jobs executed for 319–591 seconds. Replaying that run's file durations through
-the sequencer's two-worker schedule balances every shard at about 404 seconds
-of file time. The same replay with the stale history weights leaves the slowest
-shard near 547 seconds and the fastest near 293. That spread is file-schedule
-time, not runner queue, which is why each tier keeps its own history.
-CI assigns discovered files by historical duration, with a deterministic
-fallback weight for new files, rather than dividing only by file count.
-Each file keeps its full duration on one worker. Extra workers overlap other
-files in the same sequence group, and `process-boundary` runs after that
-group. History published before the competing-loader split still names only
-`javascriptLiteralPercentPaths.test.ts`. Until an artifact records
-`javascriptLiteralLoaderOwnership.test.ts`, the sequencer assigns that
-recorded duration by the pre-split case counts: 54/192 stays with the
-percent-path file and 138/192 goes to the consolidated loader file. History
-that names the retired Early, Middle, and Late shards sums those durations
-onto the consolidated file instead, and scales a partial set as three equal
-shares. That weight is the old dual-surface cost. The demoted file runs every
-loader once through the shared stdio MCP session, so the first schedule
-overestimates it. Every discovered file still belongs to exactly one shard;
+Each tier keeps its own history because stale weights from another tier, or
+from before a large test change, unbalance the shards.
+CI assigns discovered files by historical duration rather than dividing only
+by file count. A file with no recorded duration, such as a newly added or
+renamed test, is weighted at the median recorded duration until a published
+artifact records it. Each file keeps its full duration on one worker. Extra
+workers overlap other files in the same sequence group, and `process-boundary`
+runs after that group. Every discovered file still belongs to exactly one shard;
 worker limits and project ordering remain unchanged. Fast qualification keeps
 eight built shards without coverage. The merge job rejects successful reports
 with missing, unexpected, or duplicate portable files. Coverage instrumentation

knip.json

@@ -3,6 +3,7 @@
   "entry": [
     "src/main.ts",
     "src/cli.ts",
+    "src/cliOneShot.ts",
     "src/cliOutput.ts",
     "src/cliBuiltinArguments.ts",
     "src/cli/cliResultOutput.ts",

knip.production.json

@@ -3,6 +3,7 @@
   "entry": [
     "src/main.ts!",
     "src/cli.ts!",
+    "src/cliOneShot.ts!",
     "src/cliOutput.ts!",
     "src/cliBuiltinArguments.ts!",
     "src/cli/cliResultOutput.ts!",

scripts/bundle-runtime.mjs

@@ -5,7 +5,8 @@ import { fileURLToPath, pathToFileURL } from "node:url";
 
 /**
  * Reduce filesystem resolution during CLI and MCP startup. Keep dependencies
- * external, dynamic imports lazy, and original compiled worker entrypoints.
+ * external and original compiled worker entrypoints. Dynamic imports can split
+ * into lazy chunks unless the same module is also statically reachable.
  */
 export const bundleRuntime = async (root) => {
   const compiledRoot = resolve(root, "dist");
@@ -23,7 +24,7 @@ export const bundleRuntime = async (root) => {
       "cli-runtime",
       {
         cli: "cli.js",
-        cliOutput: "cliOutput.js",
+        cliOneShot: "cliOneShot.js",
         cliBuiltinArguments: "cliBuiltinArguments.js",
         cliResultOutput: "cli/cliResultOutput.js",
         mcpDoctor: "mcpDoctor.js",

scripts/ci/job-timings.mjs

@@ -1,5 +1,5 @@
 import { execFile } from "node:child_process";
-import { mkdir, readFile, writeFile } from "node:fs/promises";
+import { mkdir, writeFile } from "node:fs/promises";
 import { dirname, join, resolve } from "node:path";
 import { promisify } from "node:util";
 import { fileURLToPath } from "node:url";

scripts/ci/test-group.mjs

@@ -19,6 +19,11 @@ if (
 const child = spawn(
   process.execPath,
   [
+    // Reuse the development runner's POSIX process-group owner. Windows
+    // source tests must remain usable without the built native Job Object.
+    ...(process.platform === "win32"
+      ? []
+      : [resolve("scripts/run-exclusive.mjs"), "--unlocked", process.execPath]),
     resolve("node_modules/vitest/vitest.mjs"),
     "run",
     ...projects.flatMap(({ name }) => ["--project", name]),

scripts/generate-skill-metadata.mjs

@@ -37,21 +37,22 @@ const generated = addSkillManifestToolCount(
   FULL_CATALOG_IDENTITY.counts.mcp_tools,
 );
 if (!generated.ok) throw generated.error;
+const source = generated.value.replaceAll(latestSpecifier, pinnedSpecifier);
 const paths = await filePaths(sourceRoot);
 // Rebuild this owned output directory so removed references cannot survive a build.
 if (!check) await rm(outputRoot, { recursive: true, force: true });
 for (const relativePath of paths) {
   const path = join(outputRoot, relativePath);
   if (!check) await mkdir(dirname(path), { recursive: true });
-  const content =
-    relativePath === "SKILL.md"
-      ? generated.value
-      : await readFile(join(sourceRoot, relativePath), "utf8");
   await ensureGeneratedFile({
     path,
-    source: content
-      .replaceAll("\r\n", "\n")
-      .replaceAll(latestSpecifier, pinnedSpecifier),
+    source:
+      relativePath === "SKILL.md"
+        ? source
+        : (await readFile(join(sourceRoot, relativePath), "utf8")).replaceAll(
+            "\r\n",
+            "\n",
+          ),
     check,
     generateCommand: "npm run docs:generate",
   });

scripts/lib/artifact-e2e.d.mts

@@ -2,6 +2,17 @@ import type {
   Client,
   CallToolRequestOptions,
 } from "@modelcontextprotocol/client";
+import type { VerifierCommands } from "./verify-commands.mjs";
+
+/** Keep the opened target and its owned stdio server alive through verification. */
+export function withArtifactMcp(
+  target: string,
+  verify: (client: Client) => Promise<void>,
+  options?: {
+    readonly commands?: VerifierCommands;
+    readonly signal?: AbortSignal;
+  },
+): Promise<void>;
 
 /** Call with the server's advertised schema; callers cannot replace it. */
 export function artifactMcpCall(

scripts/lib/development-tests.d.mts

@@ -15,6 +15,7 @@ export function parseDevelopmentTestRequest(
 export function developmentTestPlan(
   request: DevelopmentTestRequest,
   baseCommit?: string,
+  resolvedPaths?: readonly string[],
 ): {
   readonly needsBuild: boolean;
   readonly artifactTasks: readonly string[];

scripts/lib/development-tests.mjs

@@ -58,7 +58,11 @@ const artifactConsumers = {
 };
 
 /** Plan source feedback or explicit tests with their runtime/artifact prerequisites. */
-export const developmentTestPlan = (request, baseCommit) => {
+export const developmentTestPlan = (
+  request,
+  baseCommit,
+  resolvedPaths = request.paths,
+) => {
   const explicit = request.paths.length > 0;
   if (
     request.mode === "changed" &&
@@ -70,12 +74,12 @@ export const developmentTestPlan = (request, baseCommit) => {
       ? []
       : SOURCE_TEST_PROJECTS.map(({ name }) => name);
   return {
-    needsBuild: request.paths.some(
+    needsBuild: resolvedPaths.some(
       (path) => projectForTest(path).group === "built",
     ),
     artifactTasks: Object.entries(artifactConsumers).flatMap(
       ([task, consumers]) =>
-        request.paths.some((path) => consumers.includes(path)) ? [task] : [],
+        resolvedPaths.some((path) => consumers.includes(path)) ? [task] : [],
     ),
     vitestArguments: [
       "run",

scripts/lib/real-ghidra-boundaries.mjs

@@ -10,7 +10,6 @@ import { access, readFile, readdir, stat } from "node:fs/promises";
 import { dirname, join } from "node:path";
 import { setTimeout as delay } from "node:timers/promises";
 import { promisify } from "node:util";
-import { GHIDRA_ANNOTATIONS_RETAINED } from "../../dist/ghidra/GhidraProviderClient.js";
 import Ajv from "ajv/dist/2020.js";
 import addFormats from "ajv-formats";
 import { OFFICIAL_TOOL_CONTRACTS } from "../../dist/contracts/officialToolContracts.js";
@@ -571,7 +570,11 @@ export async function verifyGhidraBoundaries(
   // The producer reason describes the failure; continuing-session guidance
   // is projected separately and the following live calls verify it.
   assert.match(regexError.details.reason, /stack while matching pattern/u);
-  assert.ok(regexError.message.includes(GHIDRA_ANNOTATIONS_RETAINED));
+  assert.ok(
+    regexError.message.includes(
+      "The active analysis session and annotations remain available.",
+    ),
+  );
   rejectedCalls++;
   const nestedPattern = "(".repeat(12000) + "a" + ")".repeat(12000);
   const compileFailure = await client.callTool(

scripts/lib/verifier-command-owner.mts

@@ -2,10 +2,7 @@ import { randomUUID } from "node:crypto";
 
 import type { RetainedResourceScope } from "../../src/application/AnalysisProvider.js";
 import { AnalysisCleanupLease } from "../../src/artifacts/AnalysisCleanupLease.js";
-import {
-  ArtifactResourceScope,
-  type ArtifactResourceOwner,
-} from "../../src/artifacts/ArtifactResourceScope.js";
+import { ArtifactResourceScope } from "../../src/artifacts/ArtifactResourceScope.js";
 import {
   prepareOwnedCommand,
   runOwnedCommand,
@@ -61,6 +58,7 @@ export interface VerifierCommands {
     arguments_: readonly string[],
     options?: VerifierCommandOptions,
   ): Promise<VerifierCommandOutput>;
+  /** Retain a live resource until close, independently of ordinary command execution. */
   adopt(owner: VerifierCleanupOwner, resource: string): void;
   readonly cleanupResources: readonly string[];
   close(): Promise<void>;
@@ -152,13 +150,17 @@ const commandLimits = (options: VerifierCommandOptions) => {
   };
 };
 
+type AdoptedVerifierOwner = {
+  readonly kind: "lease";
+  readonly resource: string;
+  readonly lease: AnalysisCleanupLease;
+};
+
 class VerifierCommandOwner implements VerifierCommands {
   readonly #resources = new ArtifactResourceScope();
   readonly #controller = new AbortController();
-  readonly #adopted = new WeakMap<
-    VerifierCleanupOwner,
-    ArtifactResourceOwner
-  >();
+  readonly #adopted = new WeakMap<VerifierCleanupOwner, AdoptedVerifierOwner>();
+  readonly #liveAdopted = new Set<AdoptedVerifierOwner>();
   readonly #resourceOwner: RetainedResourceScope;
   #closed = false;
   #released = false;
@@ -170,13 +172,21 @@ class VerifierCommandOwner implements VerifierCommands {
     ) => Promise<SpawnedOwnedProviderProcess>,
   ) {
     this.#resourceOwner = {
-      retain: (owner, resource) => this.adopt(owner, resource),
+      retain: (owner, resource) =>
+        this.#resources.retain(this.#adoptedOwner(owner, resource)),
       run: (work) => this.#resources.run(work),
     };
   }
 
   get cleanupResources(): readonly string[] {
-    return this.#resources.cleanupResources;
+    return [
+      ...new Set([
+        ...this.#resources.cleanupResources,
+        ...[...this.#liveAdopted].flatMap(({ lease, resource }) =>
+          lease.released ? [] : (lease.cleanupResources ?? [resource]),
+        ),
+      ]),
+    ];
   }
 
   exec(
@@ -233,6 +243,15 @@ class VerifierCommandOwner implements VerifierCommands {
   }
 
   adopt(owner: VerifierCleanupOwner, resource: string): void {
+    const retained = this.#adoptedOwner(owner, resource);
+    if (this.#closed) this.#resources.retain(retained);
+    else this.#liveAdopted.add(retained);
+  }
+
+  #adoptedOwner(
+    owner: VerifierCleanupOwner,
+    resource: string,
+  ): AdoptedVerifierOwner {
     if (this.#released)
       throw new Error("Verifier command resource release has completed");
     let retained = this.#adopted.get(owner);
@@ -244,15 +263,18 @@ class VerifierCommandOwner implements VerifierCommands {
       };
       this.#adopted.set(owner, retained);
     }
-    // An admitted command can transfer a late preparation/cleanup capability
-    // while close is joining it. Acquisition must precede this scope's release.
-    this.#resources.retain(retained);
+    return retained;
   }
 
   close(): Promise<void> {
     this.#closed = true;
     this.#controller.abort(new Error("Verifier command owner closed"));
     if (this.#closing !== undefined) return this.#closing;
+    // Live adoption owns a lifetime, not a failed cleanup. Only closure moves
+    // these leases into the retry scope; command failures enter it immediately.
+    // Late transfers during admitted work still join that same close below.
+    for (const owner of this.#liveAdopted) this.#resources.retain(owner);
+    this.#liveAdopted.clear();
     let attempt: Promise<void>;
     attempt = this.#resources
       .close()

scripts/prepack.mjs

@@ -54,7 +54,7 @@ if (await pathExists(join(root, "src"))) {
     "dist/main.js",
     "dist/mcp/main.js",
     "dist/cli-runtime/cli.js",
-    "dist/cli-runtime/cliOutput.js",
+    "dist/cli-runtime/cliOneShot.js",
     "dist/cli-runtime/cliBuiltinArguments.js",
     "dist/cli-runtime/cliResultOutput.js",
     "dist/cli-runtime/mcpDoctor.js",

scripts/rea.mjs

@@ -64,7 +64,7 @@ const runtimeFiles = [
       ? ["dist/mcp/main.js", "dist/cli-runtime/mcpDoctor.js"]
       : [
           "dist/cli-runtime/cli.js",
-          "dist/cli-runtime/cliOutput.js",
+          "dist/cli-runtime/cliOneShot.js",
           "dist/cli-runtime/cliResultOutput.js",
         ]),
 ];
@@ -104,154 +104,66 @@ if (!(await compiledRuntimeExists(runtimeFiles))) {
       );
   }
 } else {
-  const {
-    cliConfigurationError,
-    renderCliOutputArgumentError,
-    renderEmptyFilteredCliOutput,
-    renderVersionReport,
-    parseCliOutputArguments,
-  } = await import("../dist/cli-runtime/cliOutput.js");
+  const { runOneShotCli } = await import("../dist/cli-runtime/cliOneShot.js");
   guardClosedStdout();
-  const cliArgs = isShellCompletion
-    ? args
-    : builtinArguments.normalizeCliArguments(args);
-  // Completion words are transport data, including partial output flags.
-  const outputOptions = parseCliOutputArguments(
-    isShellCompletion ? [] : cliArgs,
-  );
-  const outputArguments = outputOptions.validation;
-  // `--version` with an explicit format reports the runtime as a document;
-  // plain `--version` stays Incur's bare version string.
-  const versionOutput = renderVersionReport(
-    outputOptions,
-    {
-      packageName: packageJson.name,
-      packageVersion: packageJson.version,
-      versions: process.versions,
-      platform: process.platform,
-      architecture: process.arch,
-    },
-    process.stdout.isTTY === true,
-  );
-  const configurationError = cliConfigurationError(process.env);
   let dispatchFailure;
   let closeCliResources;
   // Incur can exit from the callback below before this function returns.
   // That callback owns the status, including a closed-pipe exit of 141.
   let exiting = false;
   try {
-    // Doctor is the readiness report for this failure. Other commands still stop
-    // here so they do not run with an unusable log level.
-    if (
-      configurationError !== undefined &&
-      (isShellCompletion ||
-        builtinArguments.splitIncurBuiltinArguments(cliArgs).command[0] !==
-          "doctor")
-    ) {
-      process.stderr.write(`${configurationError}\n`);
-      process.exitCode = 1;
-    } else if (!outputArguments.ok) {
-      writeStdout(renderCliOutputArgumentError(outputArguments));
-      process.exitCode = 1;
-    } else if (versionOutput !== undefined) {
-      writeStdout(versionOutput);
-    } else if (
-      !isShellCompletion &&
-      builtinArguments.isProjectScopedAmpRegistration(cliArgs)
-    ) {
-      // Amp has global settings only. Share the parsed scope fact with the
-      // registration workflow and refuse before loading the command runtime.
-      //
-      // A caller that asked for a structured document gets the refusal in
-      // that document, through the same pre-execution error renderer the
-      // output-argument branch uses; anything else keeps the plain message.
-      const refusalMessage =
-        "Amp does not support project-scoped registration. Run without --no-global to register Amp globally, or select a client with project settings. No client settings were changed.";
-      const refusalFormat = outputOptions.format;
-      if (
-        refusalFormat === "json" ||
-        refusalFormat === "jsonl" ||
-        refusalFormat === "yaml"
-      ) {
-        writeStdout(
-          renderCliOutputArgumentError({
-            ok: false,
-            format: refusalFormat,
-            code: "AMP_PROJECT_SCOPE_UNSUPPORTED",
-            message: refusalMessage,
-          }),
-        );
-      } else {
-        process.stderr.write(`${refusalMessage}\n`);
-      }
-      process.exitCode = 1;
-    } else {
-      const cliRuntime = await import("../dist/cli-runtime/cli.js");
-      closeCliResources = cliRuntime.closeCliResources;
-      const { createCli } = cliRuntime;
-      const { createCliResultOutput } =
-        await import("../dist/cli-runtime/cliResultOutput.js");
-      let wroteOutput = false;
-      const resultOutput = createCliResultOutput(outputOptions, process.stdout);
-      try {
-        await createCli(process.env, resultOutput).serve(cliArgs, {
-          stdout: (output) => {
-            if (resultOutput?.handled) {
-              if (resultOutput.failed && !stdoutClosed)
-                process.stderr.write(output);
-              return;
-            }
-            if (output.length > 0) wroteOutput = true;
-            if (resultOutput?.captureSdkOutput(output)) return;
-            writeStdout(output);
-          },
-          // A command failure caused by the closed pipe keeps the pipe's status.
-          // Worker shutdown is shared with the dispatch finally below. A failed
-          // shutdown is reported here because this callback is what exits.
-          exit: (code) => {
-            exiting = true;
-            void (async () => {
-              let status = typeof code === "number" ? code : 0;
-              try {
-                await closeCliResources();
-              } catch (cleanup) {
-                process.stderr.write(
-                  `Failed to close CLI resources: ${cleanupFailureText(cleanup)}\n`,
-                );
-                if (status === 0) status = 1;
-              }
-              try {
-                await resultOutput?.finishSdkOutput(status);
-              } catch (cause) {
-                if (!stdoutClosed)
-                  process.stderr.write(`${cleanupFailureText(cause)}\n`);
-                if (status === 0) status = 1;
-              }
-              process.exit(stdoutClosed ? SIGPIPE_EXIT_STATUS : status);
-            })();
-          },
-        });
-      } catch (cause) {
-        if (wroteOutput) throw cause;
-        const recovered = await resultOutput
-          ?.recoverFormattingFailure(cause)
-          .catch((writeFailure) => {
-            // Recovery writes outside Incur's command boundary. A reader that
-            // declined this document keeps the same quiet SIGPIPE outcome.
-            if (!stdoutClosed) throw writeFailure;
-            return true;
-          });
-        if (!recovered) throw cause;
-        process.exitCode = stdoutClosed ? SIGPIPE_EXIT_STATUS : 1;
-      }
-      if (!wroteOutput && !resultOutput?.handled)
-        writeStdout(renderEmptyFilteredCliOutput(outputOptions) ?? "");
-      // An SDK failure exits through the callback, which owns its pending write.
-      if (!exiting)
-        await resultOutput?.finishSdkOutput(
-          typeof process.exitCode === "number" ? process.exitCode : 0,
-        );
-    }
+    await runOneShotCli(args, {
+      environment: process.env,
+      runtime: {
+        packageName: packageJson.name,
+        packageVersion: packageJson.version,
+        versions: process.versions,
+        platform: process.platform,
+        architecture: process.arch,
+      },
+      stdoutIsTerminal: process.stdout.isTTY === true,
+      resultDestination: process.stdout,
+      closedStdoutStatus: SIGPIPE_EXIT_STATUS,
+      writeStdout,
+      writeStderr: (text) => {
+        process.stderr.write(text);
+      },
+      stdoutClosed: () => stdoutClosed,
+      exitCode: () =>
+        typeof process.exitCode === "number" ? process.exitCode : 0,
+      setExitCode: (code) => {
+        process.exitCode = code;
+      },
+      ownCliResources: (close) => {
+        closeCliResources = close;
+      },
+      exitRequested: () => exiting,
+      // A command failure caused by the closed pipe keeps the pipe's status.
+      // Worker shutdown is shared with the dispatch finally below. A failed
+      // shutdown is reported here because this callback is what exits.
+      exit: (code, finishOutput) => {
+        exiting = true;
+        void (async () => {
+          let status = code;
+          try {
+            await closeCliResources();
+          } catch (cleanup) {
+            process.stderr.write(
+              `Failed to close CLI resources: ${cleanupFailureText(cleanup)}\n`,
+            );
+            if (status === 0) status = 1;
+          }
+          try {
+            await finishOutput(status);
+          } catch (cause) {
+            if (!stdoutClosed)
+              process.stderr.write(`${cleanupFailureText(cause)}\n`);
+            if (status === 0) status = 1;
+          }
+          process.exit(stdoutClosed ? SIGPIPE_EXIT_STATUS : status);
+        })();
+      },
+    });
   } catch (cause) {
     if (cause?.code === "EPIPE") {
       stdoutClosed = true;

scripts/run-development-tests.mjs

@@ -1,12 +1,13 @@
 import { spawn } from "node:child_process";
-import { stat } from "node:fs/promises";
-import { resolve } from "node:path";
+import { realpath, stat } from "node:fs/promises";
+import { relative, resolve, sep } from "node:path";
 import { fileURLToPath } from "node:url";
 
 import {
   developmentTestPlan,
   parseDevelopmentTestRequest,
 } from "./lib/development-tests.mjs";
+import { projectForTest } from "./lib/test-projects.mjs";
 
 let child;
 let forceTimer;
@@ -29,9 +30,22 @@ for (const signal of Object.keys(signalExitCodes))
 try {
   const [mode, ...arguments_] = process.argv.slice(2);
   const request = parseDevelopmentTestRequest(mode, arguments_);
+  const resolvedPaths = [];
+  const repository = await realpath(".");
   for (const path of request.paths) {
     if (!(await stat(path)).isFile())
       throw new Error(`Test path is not a file: ${path}`);
+    // Ownership follows the filesystem's canonical spelling. Execution keeps
+    // the caller path so exact discovery still rejects excluded symlink aliases.
+    const resolvedPath = relative(repository, await realpath(path))
+      .split(sep)
+      .join("/");
+    if (
+      request.mode !== "focused" &&
+      projectForTest(resolvedPath).group !== "source"
+    )
+      throw new Error(`Use test:focused for built test projects: ${path}`);
+    resolvedPaths.push(resolvedPath);
   }
   let baseCommit;
   if (request.mode === "changed") {
@@ -42,7 +56,7 @@ try {
       );
     baseCommit = base.output.trim();
   }
-  const plan = developmentTestPlan(request, baseCommit);
+  const plan = developmentTestPlan(request, baseCommit, resolvedPaths);
   if (request.dryRun) {
     console.log(JSON.stringify({ ...request, baseCommit, ...plan }, null, 2));
   } else {

scripts/verify-go-binary.mjs

@@ -1,22 +1,23 @@
 #!/usr/bin/env node
 import assert from "node:assert/strict";
 import { isUtf8 } from "node:buffer";
-import { execFile } from "node:child_process";
 import { createHash } from "node:crypto";
 import { lstat, mkdir, readFile, writeFile } from "node:fs/promises";
 import { isAbsolute, join } from "node:path";
 import { fileURLToPath } from "node:url";
-import { promisify } from "node:util";
 import { Client } from "@modelcontextprotocol/client";
-import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
 import { parseEvidence } from "../dist/domain/evidence.js";
 import { goBinarySchema } from "../dist/domain/go/goBinary.js";
 import { PrivateRuntimeRoot } from "../dist/process/PrivateRuntimeRoot.js";
+import { AnalysisCleanupLease } from "../dist/artifacts/AnalysisCleanupLease.js";
 import { artifactMcpCall } from "./lib/artifact-e2e.mjs";
 import { requireMcpOutput } from "./lib/mcp-verifier-results.mjs";
 import { createVerifierRun, completeVerifierRun } from "./lib/verifier-run.mjs";
+import { createVerifierCommands } from "./lib/verify-commands.mjs";
+import { createOwnedMcpStdioTransport } from "./lib/owned-mcp-stdio.mjs";
+import { closeVerifierResources } from "./lib/verifier-cleanup.mjs";
 
-const execute = promisify(execFile);
+const commands = await createVerifierCommands();
 const compiler = process.env.REA_VERIFY_GO_COMMAND ?? "go";
 if (process.env.REA_VERIFY_GO_COMMAND !== undefined && !isAbsolute(compiler))
   throw new Error(
@@ -51,13 +52,16 @@ const goEnvironment = {
   // Go's own verifier seam isolates telemetry; no global configuration is changed.
   TEST_TELEMETRY_DIR: join(root.path, "telemetry"),
 };
-const client = new Client({ name: "go-binary-verifier", version: "1" });
-const transport = new StdioClientTransport({
-  command: process.execPath,
-  args: [entrypoint, "mcp"],
-  env: environment,
-  stderr: "pipe",
-});
+let client;
+let transport;
+const cleanup = new AnalysisCleanupLease(async () => {
+  await closeVerifierResources([
+    () => client?.close(),
+    () => transport?.close(),
+    () => commands.close(),
+  ]);
+  await root.close();
+}, [root.path]);
 const targets = [
   ["linux", "amd64", "elf", 64, "little"],
   ["linux", "386", "elf", 32, "little"],
@@ -73,10 +77,19 @@ const verified = [];
 const omittedCoverage = [];
 const failures = [];
 try {
+  client = new Client({ name: "go-binary-verifier", version: "1" });
+  transport = await createOwnedMcpStdioTransport({
+    command: process.execPath,
+    arguments: [entrypoint, "mcp"],
+    hostEnvironment: environment,
+    cwd: process.cwd(),
+    expectedCommand: null,
+    sweepTokenOwnedProcesses: true,
+  });
   await prepareFixture();
   try {
     compilerVersion = (
-      await execute(compiler, ["version"], {
+      await commands.exec(compiler, ["version"], {
         cwd: root.path,
         env: goEnvironment,
         timeout: 10_000,
@@ -185,7 +198,7 @@ try {
     const path = join(root.path, "linux-amd64-byte-strings");
     const expected = JSON.parse(
       (
-        await execute(oracle, ["--build-byte-fixture", compiler, path], {
+        await commands.exec(oracle, ["--build-byte-fixture", compiler, path], {
           cwd: sourceRoot,
           env: { ...goEnvironment, GOOS: "linux", GOARCH: "amd64" },
           timeout: 120_000,
@@ -198,21 +211,16 @@ try {
 } catch (cause) {
   failures.push(cause);
 } finally {
-  for (const close of [
-    () => client.close(),
-    () => transport.close(),
-    () => root.close(),
-  ]) {
-    try {
-      await close();
-    } catch (cause) {
-      failures.push(cause);
-    }
+  try {
+    await cleanup.release();
+  } catch (cause) {
+    failures.push(cause);
   }
 }
 const verifier = await completeVerifierRun(run);
 try {
-  await assert.rejects(lstat(root.path), { code: "ENOENT" });
+  if (cleanup.released)
+    await assert.rejects(lstat(root.path), { code: "ENOENT" });
   if (verifier.process_lineage.status === "verified")
     assert.deepEqual(verifier.process_lineage.descendants, []);
 } catch (cause) {
@@ -226,15 +234,21 @@ if (failures.length > 0) {
         compiler: compilerVersion,
         verified,
         omitted_coverage: omittedCoverage,
+        retained_root: cleanup.released ? null : root.path,
         verifier,
       },
       null,
       2,
     ),
   );
-  throw new AggregateError(
-    failures,
-    "Real Go metadata verification failed; cleanup failures are retained.",
+  throw Object.assign(
+    new AggregateError(
+      failures,
+      cleanup.released
+        ? "Real Go metadata verification failed; cleanup failures are retained."
+        : `Real Go metadata verification failed; cleanup is incomplete at ${root.path}.`,
+    ),
+    { cleanupLeases: cleanup.released ? [] : [cleanup] },
   );
 }
 console.log(
@@ -256,7 +270,7 @@ console.log(
 );
 
 async function go(args, overrides = {}) {
-  return execute(compiler, args, {
+  return commands.exec(compiler, args, {
     cwd: sourceRoot,
     env: { ...goEnvironment, ...overrides },
     timeout: 120_000,
@@ -356,7 +370,7 @@ async function inspect(mode, path, includeStringLiterals = false) {
       ...(includeStringLiterals ? ["--include-string-literals"] : []),
       "--json",
     ];
-    const { stdout } = await execute(process.execPath, args, {
+    const { stdout } = await commands.exec(process.execPath, args, {
       env: environment,
       timeout: 30_000,
       maxBuffer: 16 * 1024 * 1024,
@@ -424,7 +438,7 @@ async function verify(
     byteExpected ??
     JSON.parse(
       (
-        await execute(oracle, [oraclePath], {
+        await commands.exec(oracle, [oraclePath], {
           timeout: 10_000,
           maxBuffer: 1024 * 1024,
         })

scripts/verify-package-setup.mjs

@@ -1,4 +1,3 @@
-import { deepStrictEqual } from "node:assert";
 import {
   lstat,
   mkdir,
@@ -13,7 +12,6 @@ import { dirname, isAbsolute, join } from "node:path";
 import { parse as parseToml } from "smol-toml";
 
 import { TOOL_CONTRACTS } from "../dist/contracts/toolContracts.js";
-import { PRODUCT_IDENTITY } from "../dist/identity.js";
 import { pathExists } from "./lib/verify-package-core.mjs";
 import { parse as parseJsonc } from "jsonc-parser";
 import { skillReferenceIssues } from "./lib/docs-facts.mjs";

scripts/verify-published-package.mjs

@@ -99,8 +99,7 @@ async function verifyPublishedUpdate({
     environment,
   );
 
-  await commands.exec(
-    "npm",
+  await commands.execNpm(
     [
       "install",
       "--global",
@@ -118,7 +117,7 @@ async function verifyPublishedUpdate({
   );
 
   const npmRoot = (
-    await commands.exec("npm", ["root", "--global"], {
+    await commands.execNpm(["root", "--global"], {
       cwd: canaryRoot,
       env: environment,
       timeout: 30000,
@@ -186,8 +185,7 @@ if (result.status === "failed") process.exitCode = 1;
 }
 
 async function findPreviousPublishedVersion(targetVersion, cwd, environment) {
-  const { stdout } = await commands.exec(
-    "npm",
+  const { stdout } = await commands.execNpm(
     ["view", "rea-agents", "versions", "--json"],
     { cwd, env: environment, timeout: 30000, maxBuffer: 16 * 1024 * 1024 },
   );

scripts/verify-real-ghidra-dol.mjs

@@ -1,7 +1,6 @@
 #!/usr/bin/env node
 import assert from "node:assert/strict";
 import { execFile } from "node:child_process";
-import { createHash } from "node:crypto";
 import { readFile, mkdtemp, mkdir, rm } from "node:fs/promises";
 import { tmpdir } from "node:os";
 import { join } from "node:path";
@@ -13,7 +12,6 @@ import { parseBinaryTarget } from "../dist/application/BinaryTargetResolver.js";
 import { ArtifactResourceScope } from "../dist/artifacts/ArtifactResourceScope.js";
 import { readGhidraLanguageCatalog } from "../dist/ghidra/GhidraLanguageCatalog.js";
 import { parseEvidence } from "../dist/domain/evidence.js";
-import { parseAnalysisSnapshot } from "../dist/domain/analysisSnapshot.js";
 import { functionDossierSchema } from "../dist/domain/native/functionDossier.js";
 import { inspectVerifierGhidraInstallation } from "./lib/ghidra-installation-inspection.mjs";
 import { buildGameCubeDolFixture } from "../tests/conformance/ghidra/dol-fixture.mjs";
@@ -96,8 +94,6 @@ async function languageInstalled(languageId) {
   }
 }
 
-const sha256 = (bytes) => createHash("sha256").update(bytes).digest("hex");
-
 /** One MCP session plus CLI check. `languageOverride` is set only for the Gekko pass. */
 async function runPass({ fixture, languageOverride, label, layout }) {
   const env = { ...baseEnv };

scripts/verify/android/failure-recovery.mjs

@@ -151,7 +151,7 @@ export async function verifyAndroidFailureRecovery({
             failure_code: "runtime_unavailable",
           },
           availabilityReason: "provider_unavailable",
-          availabilityRemediation: /selected Java runtime/u,
+          availabilityRemediation: /selected Java launcher/u,
           message: /selected Java runtime/u,
           remediation: /selected launcher/u,
         });
@@ -194,12 +194,7 @@ export async function verifyAndroidFailureRecovery({
         repository,
         available: false,
         unavailableReason: scenario.availabilityReason ?? "provider_missing",
-        remediation:
-          scenario.availabilityRemediation ??
-          new RegExp(
-            scenario.observation.replace(/[.*+?^${}()|[\]\\]/gu, "\\$&"),
-            "u",
-          ),
+        remediation: scenario.availabilityRemediation ?? scenario.remediation,
       });
       console.log(`PASS CLI applicable recovery: ${scenario.label}`);
     }

scripts/verify/rea-numeric-limits.mjs

@@ -94,29 +94,36 @@ const stringBindings = (sourceFile, symbols, unresolved) => {
     ts.forEachChild(node, collect);
   };
   collect(sourceFile);
-  const potential = (node) => {
+  const potential = (node, matchesCandidate) => {
     const expression = unwrapExpression(node);
     if (exactLimit(literalText(expression)) !== undefined) return true;
-    if (ts.isIdentifier(expression)) return candidates.has(expression.text);
+    if (ts.isIdentifier(expression)) return matchesCandidate(expression);
     if (ts.isPropertyAccessExpression(expression))
       return (
         exactLimit(expression.name.text) !== undefined ||
-        potential(expression.expression)
+        potential(expression.expression, matchesCandidate)
       );
     if (ts.isElementAccessExpression(expression))
       return (
         expression.argumentExpression !== undefined &&
-        potential(expression.argumentExpression)
+        potential(expression.argumentExpression, matchesCandidate)
       );
     if (ts.isObjectLiteralExpression(expression))
       return expression.properties.some(
         (property) =>
-          ts.isPropertyAssignment(property) && potential(property.initializer),
+          ts.isPropertyAssignment(property) &&
+          potential(property.initializer, matchesCandidate),
       );
     if (ts.isConditionalExpression(expression))
-      return potential(expression.whenTrue) || potential(expression.whenFalse);
+      return (
+        potential(expression.whenTrue, matchesCandidate) ||
+        potential(expression.whenFalse, matchesCandidate)
+      );
     if (ts.isBinaryExpression(expression))
-      return potential(expression.left) || potential(expression.right);
+      return (
+        potential(expression.left, matchesCandidate) ||
+        potential(expression.right, matchesCandidate)
+      );
     return false;
   };
   let changed = true;
@@ -125,13 +132,34 @@ const stringBindings = (sourceFile, symbols, unresolved) => {
     for (const declaration of declarations) {
       if (
         !candidates.has(declaration.name.text) &&
-        potential(declaration.initializer)
+        potential(declaration.initializer, (identifier) =>
+          candidates.has(identifier.text),
+        )
       ) {
         candidates.add(declaration.name.text);
         changed = true;
       }
     }
   }
+  // Names narrow the search cheaply, but only the lexical declaration can
+  // establish that a particular use is an alias of a limit.
+  const lexicalCandidates = new Set();
+  changed = true;
+  while (changed) {
+    changed = false;
+    for (const declaration of declarations) {
+      if (
+        candidates.has(declaration.name.text) &&
+        !lexicalCandidates.has(declaration) &&
+        potential(declaration.initializer, (identifier) =>
+          lexicalCandidates.has(symbols(identifier)?.valueDeclaration),
+        )
+      ) {
+        lexicalCandidates.add(declaration);
+        changed = true;
+      }
+    }
+  }
   const unknown = (node, reason) => {
     const path = sourceFile.fileName;
     const offset = node.getStart();
@@ -152,11 +180,12 @@ const stringBindings = (sourceFile, symbols, unresolved) => {
       declaration === undefined ||
       !ts.isVariableDeclaration(declaration) ||
       !ts.isVariableDeclarationList(declaration.parent) ||
-      declaration.initializer === undefined
+      declaration.initializer === undefined ||
+      !lexicalCandidates.has(declaration)
     )
       return undefined;
     if ((declaration.parent.flags & ts.NodeFlags.Const) === 0) {
-      unknown(identifier, "mutable setting-key or value alias");
+      unknown(declaration.name, "mutable setting-key or value alias");
       return undefined;
     }
     if (declaration.pos >= identifier.pos) {
@@ -890,7 +919,7 @@ if (direct) {
   const entries = parseNumericLimitRegistry(registry);
   const listed = new Map();
   for (const entry of entries) {
-    const { name, policy, read, reason } = entry;
+    const { name, read, reason } = entry;
     if (listed.has(name)) failures.push(`${name} is listed twice`);
     listed.set(name, entry);
     if (reason.trim().length < 40)

src/adb/AdbCommand.capture.test.ts

@@ -20,6 +20,7 @@ it.each([
     source:
       'process.stdout.write("partial-雪"); process.stderr.write("waiting\\n"); setInterval(() => {}, 1000)',
     code: "provider_timeout",
+    timeoutMs: 500,
     stdout: "partial-雪",
     stderr: "waiting\n",
   },
@@ -28,6 +29,7 @@ it.each([
     source:
       'process.stdout.write("partial-雪"); process.stderr.write("adb: error: device offline\\n"); process.exitCode = 1',
     code: "unsupported_target",
+    expectedPath: "caller-device",
     stdout: "partial-雪",
     stderr: "adb: error: device offline\n",
   },
@@ -36,6 +38,7 @@ it.each([
     source:
       'process.stdout.write("partial-雪"); process.stderr.write("adb: error: cannot stat \'/host/token file\': Permission denied\\n"); process.exitCode = 1',
     code: "unsupported_target",
+    expectedPath: "/host/token file",
     stdout: "partial-雪",
     stderr: "adb: error: cannot stat '/host/token file': Permission denied\n",
   },
@@ -54,7 +57,7 @@ it.each([
     },
     ["-e", testCase.source],
     {
-      timeoutMs: testCase.name === "timeout" ? 500 : 5_000,
+      timeoutMs: testCase.timeoutMs ?? 5_000,
       maxOutputBytes: testCase.maxOutputBytes ?? 1024,
     },
     undefined,
@@ -75,12 +78,34 @@ it.each([
     },
   });
   if (testCase.code === "unsupported_target")
-    expect(projected.details?.path).toBe(
-      testCase.name === "local path refusal"
-        ? "/host/token file"
-        : "caller-device",
-    );
+    expect(projected.details?.path).toBe(testCase.expectedPath);
+  if (testCase.code === "provider_timeout") {
+    expect(projected.details?.timeout_ms).toBe(testCase.timeoutMs);
+    expect(projected.details?.limit).toBeUndefined();
+  }
   expect(JSON.stringify(projected)).not.toContain(
     "ambient-value-must-not-be-persisted",
   );
 });
+
+it.each([true, false])(
+  "projects a truncated adb capture as retryable only for volatile output (%s)",
+  async (volatileOutput) => {
+    const result = await runAdbCapture(
+      volatileOutput ? "read_adb_logcat" : "install_adb_package",
+      { device: "device" },
+      { binary: process.execPath, pathSource: "environment", environment: {} },
+      ["-e", 'process.stdout.write("X".repeat(4096))'],
+      { timeoutMs: 5_000, maxOutputBytes: 1024, volatileOutput },
+      undefined,
+    );
+    if (result.ok) throw new Error("Expected a capture allowance refusal");
+    const projected = projectAnalysisError(result.error);
+    expect(projected.retryable).toBe(volatileOutput);
+    expect(projected).toMatchObject({
+      code: "resource_constraint",
+      details: { reported_limits: { max_output_bytes: 1024 } },
+    });
+    expect(projected.details?.limit).toBeUndefined();
+  },
+);

src/adb/AdbCommand.test.ts

@@ -1,5 +1,4 @@
 import { expect, it, vi } from "vitest";
-import { projectAnalysisError } from "../domain/analysisErrorProjection.js";
 import {
   ExecFileOutputCancelledError,
   execFileBytes,
@@ -212,65 +211,6 @@ it("does not infer cancellation from a foreign ABORT_ERR code", async () => {
   });
 });
 
-it.each([true, false])(
-  "projects a truncated adb capture as retryable only for volatile output (%s)",
-  async (volatileOutput) => {
-    vi.mocked(execFileBytes).mockRejectedValueOnce(
-      Object.assign(new Error("too much"), {
-        code: 1,
-        stdout: "",
-        stderr: "",
-        signal: null,
-        killed: false,
-        outputTruncated: true,
-      }),
-    );
-    const result = await runAdbCapture(
-      volatileOutput ? "read_adb_logcat" : "install_adb_package",
-      { device: "device" },
-      { binary: "/host/adb", pathSource: "environment", environment: {} },
-      volatileOutput ? ["logcat", "-d"] : ["install", "app.apk"],
-      { timeoutMs: 100, maxOutputBytes: 1024, volatileOutput },
-      undefined,
-    );
-    if (result.ok) throw new Error("Expected a capture allowance refusal");
-    const projected = projectAnalysisError(result.error);
-    expect(projected.retryable).toBe(volatileOutput);
-    expect(projected).toMatchObject({
-      code: "resource_constraint",
-      details: { reported_limits: { max_output_bytes: 1024 } },
-    });
-    expect(projected.details?.limit).toBeUndefined();
-  },
-);
-
-it("reports the effective fixed adb deadline", async () => {
-  vi.mocked(execFileBytes).mockRejectedValueOnce(
-    Object.assign(new Error("deadline"), {
-      code: "ETIMEDOUT",
-      stdout: "",
-      stderr: "",
-      signal: null,
-      killed: true,
-    }),
-  );
-  const result = await runAdbCapture(
-    "pull_adb_file",
-    { device: "device" },
-    { binary: "/host/adb", pathSource: "environment", environment: {} },
-    ["pull", "/remote"],
-    { timeoutMs: 600_000, maxOutputBytes: 1024 },
-    undefined,
-  );
-  if (result.ok) throw new Error("Expected an adb deadline");
-  const projected = projectAnalysisError(result.error);
-  expect(projected).toMatchObject({
-    code: "provider_timeout",
-    details: { timeout_ms: 600_000 },
-  });
-  expect(projected.details?.limit).toBeUndefined();
-});
-
 it("keeps a physical failure even when the later abort reason is that same error", async () => {
   const controller = new AbortController();
   const failure = new Error("physical command failure");

src/adb/AdbProvider.test.ts

@@ -191,7 +191,7 @@ fi`,
     await Promise.allSettled([inspection, closing]);
     await provider.close();
   }
-}, 15_000);
+});
 
 it("fails closed when adb exits successfully without a recognized version", async () => {
   const stdout = `unexpected adb output ${"x".repeat(700)}\n`;

src/adb/AdbProvider.ts

@@ -84,6 +84,7 @@ import {
 import { RegularFileOperationCancelled } from "../filesystem/RegularFile.js";
 import { OwnedFileHandle } from "../filesystem/OwnedFileHandle.js";
 import { createEvidence } from "../domain/evidence.js";
+import { createAnalysisExecutionEvidence } from "../domain/analysisObservation.js";
 import {
   PrivateRuntimeRoot,
   type PrivateRuntimeRootOptions,
@@ -993,15 +994,12 @@ export class AdbProvider {
     const partialObservation =
       execution === undefined
         ? undefined
-        : createEvidence(execution.subject ?? undefined, execution.provider, {
+        : createAnalysisExecutionEvidence(
             operation,
             parameters,
-            result: execution.result,
-            rawResult: execution.rawResult,
-            limitations: execution.limitations,
-            locations: execution.locations,
-            confidence: "observed",
-          });
+            execution,
+            "observed",
+          );
     return new ProviderCleanupError(
       ADB_PROVIDER_IDENTITY.id,
       resources,

src/adb/AdbVersionProbe.test.ts

@@ -47,19 +47,32 @@ it("reports the fixed adb version-probe deadline", async () => {
 });
 
 it("reports the fixed adb version-output cap", async () => {
-  vi.mocked(execFileOutput).mockRejectedValueOnce(
-    Object.assign(new Error("too much"), {
-      code: "ERR_CHILD_PROCESS_STDIO_MAXBUFFER",
-      stdout: "",
-      stderr: "",
-      signal: null,
-      killed: false,
-      outputTruncated: true,
-    }),
-  );
-  const failure = await inspectAdbClient(selection).catch(
-    (cause: unknown) => cause,
-  );
+  let selected: string = selection.binary;
+  if (process.platform === "win32") {
+    vi.mocked(execFileOutput).mockRejectedValueOnce(
+      Object.assign(new Error("too much"), {
+        code: "ERR_CHILD_PROCESS_STDIO_MAXBUFFER",
+        stdout: "",
+        stderr: "",
+        signal: null,
+        killed: false,
+        outputTruncated: true,
+      }),
+    );
+  } else {
+    const root = await mkdtemp(join(tmpdir(), "rea-adb-version-cap-"));
+    onTestFinished(() => rm(root, { recursive: true, force: true }));
+    selected = join(root, "adb");
+    await writeFile(
+      selected,
+      `#!${process.execPath}\nprocess.stdout.write("x".repeat(100_000));\n`,
+      { mode: 0o755 },
+    );
+  }
+  const failure = await inspectAdbClient({
+    ...selection,
+    binary: selected,
+  }).catch((cause: unknown) => cause);
   const projected = projectAnalysisError(
     failure as Parameters<typeof projectAnalysisError>[0],
   );
@@ -74,25 +87,28 @@ it("reports the fixed adb version-output cap", async () => {
   expect(projected.details?.limit).toBeUndefined();
 });
 
-it("distinguishes execute denial from missing or unreadable version output", async () => {
-  vi.mocked(execFileOutput).mockRejectedValueOnce(
-    Object.assign(new Error("denied"), {
-      code: "EACCES",
-      spawnFailed: true,
-      stdout: "",
-      stderr: "",
-      signal: null,
-      killed: false,
-    }),
-  );
-  await expect(inspectAdbClient(selection)).rejects.toMatchObject({
-    code: "binary_not_executable",
-    diagnostics: {
-      selected_binary: selection.binary,
-      filesystem_error: "EACCES",
-    },
-  });
-});
+it.skipIf(process.platform !== "win32")(
+  "classifies an EACCES execute refusal on Windows",
+  async () => {
+    vi.mocked(execFileOutput).mockRejectedValueOnce(
+      Object.assign(new Error("denied"), {
+        code: "EACCES",
+        spawnFailed: true,
+        stdout: "",
+        stderr: "",
+        signal: null,
+        killed: false,
+      }),
+    );
+    await expect(inspectAdbClient(selection)).rejects.toMatchObject({
+      code: "binary_not_executable",
+      diagnostics: {
+        selected_binary: selection.binary,
+        filesystem_error: "EACCES",
+      },
+    });
+  },
+);
 
 it("names an EPERM execute refusal separately from missing permission", async () => {
   vi.mocked(execFileOutput).mockRejectedValueOnce(

src/android/AndroidTargetSnapshot.ts

@@ -58,7 +58,7 @@ export const androidSnapshotCopy = (
     : undefined;
 
 /** Adopt a snapshot descriptor whose close rejected, keyed by its exact path. */
-export type AndroidSnapshotCloseRetention = (
+type AndroidSnapshotCloseRetention = (
   path: string,
   owner: OwnedFileHandle,
 ) => void;
@@ -69,8 +69,8 @@ export const snapshotAndroidTarget = async (
   sha256: string,
   root: string,
   operation: AndroidOperation,
-  signal?: AbortSignal,
-  retainFailedClose?: AndroidSnapshotCloseRetention,
+  signal: AbortSignal | undefined,
+  retainFailedClose: AndroidSnapshotCloseRetention,
 ): Promise<string> => {
   const snapshot = join(root, "target.apk");
   try {
@@ -105,8 +105,8 @@ export const snapshotAndroidEngine = async (
   sha256: string,
   root: string,
   operation: AndroidOperation,
-  signal?: AbortSignal,
-  retainFailedClose?: AndroidSnapshotCloseRetention,
+  signal: AbortSignal | undefined,
+  retainFailedClose: AndroidSnapshotCloseRetention,
 ): Promise<{ path: string; sha256: string }> => {
   const path = join(root, "engine.jar");
   try {
@@ -159,17 +159,14 @@ const copyAndroidSnapshot = async (
   snapshotPath: string,
   expectedSha256: string,
   signal: AbortSignal | undefined,
-  retainFailedClose: AndroidSnapshotCloseRetention | undefined,
+  retainFailedClose: AndroidSnapshotCloseRetention,
 ): Promise<AndroidSnapshotCopy> => {
   assertRegularFileActive(signal);
   let removeSnapshot = false;
-  const retain = (path: string) =>
-    retainFailedClose === undefined
-      ? {}
-      : {
-          retainFailedClose: (owner: OwnedFileHandle) =>
-            retainFailedClose(path, owner),
-        };
+  const retain = (path: string) => ({
+    retainFailedClose: (owner: OwnedFileHandle) =>
+      retainFailedClose(path, owner),
+  });
   try {
     // Close each descriptor through its owner. A rejected close keeps the
     // copy outcome and stays retained until that owner is retried; the
@@ -287,7 +284,7 @@ const openSnapshotSource = async (
   path: string,
   signal: AbortSignal | undefined,
   retention: {
-    readonly retainFailedClose?: (owner: OwnedFileHandle) => void;
+    readonly retainFailedClose: (owner: OwnedFileHandle) => void;
   },
 ): Promise<StableRegularFileDescriptor> => {
   try {
@@ -305,10 +302,9 @@ const openSnapshotSource = async (
 };
 
 const retainAdmissionOwner = (
-  retain: ((owner: OwnedFileHandle) => void) | undefined,
+  retain: (owner: OwnedFileHandle) => void,
   owner: OwnedFileHandle,
 ): RegularFileCleanupFailure["retention"] => {
-  if (retain === undefined) return { kind: "not_requested" };
   try {
     retain(owner);
     return { kind: "callback_returned" };

src/android/JadxProvider.ts

@@ -313,10 +313,10 @@ export class JadxProvider implements AndroidAnalysisPort {
           : AbortSignal.any([signal, this.#shutdown.signal]),
     });
     this.#inspections.add(pending);
-    void pending.then(
-      () => this.#inspections.delete(pending),
-      () => this.#inspections.delete(pending),
-    );
+    const remove = (): void => {
+      this.#inspections.delete(pending);
+    };
+    void pending.then(remove, remove);
     return pending;
   }
 

src/apkSignature/ApkSignatureProvider.ts

@@ -415,11 +415,11 @@ export class ApkSignatureProvider implements ApkSignaturePort {
     const { v1, decoded: v1Decoded, duplicates } = v1Read.value;
 
     const v2 = decodeScheme(block.value, APK_BLOCK_ID.signatureV2, "v2");
-    if (!v2.ok) return err(v2.error);
+    if (!v2.ok) return v2;
     const v3 = decodeScheme(block.value, APK_BLOCK_ID.signatureV3, "v3");
-    if (!v3.ok) return err(v3.error);
+    if (!v3.ok) return v3;
     const v31 = decodeScheme(block.value, APK_BLOCK_ID.signatureV31, "v3.1");
-    if (!v31.ok) return err(v31.error);
+    if (!v31.ok) return v31;
 
     // Account for representation expansion before allocating the report.
     const projected = projectedOutputBytes(

src/apktool/ApktoolConfiguration.test.ts

@@ -1,5 +1,6 @@
 import * as commandOwner from "../process/OwnedCommand.js";
 import { failedInspectionDescriptor } from "../../tests/fixtures/processInspectionCleanup.js";
+import { awaitAcknowledgement } from "../../tests/fixtures/lifecycleAcknowledgement.js";
 import {
   AnalysisError,
   primaryAnalysisError,
@@ -486,6 +487,7 @@ it("cancels a pending filesystem lookup and owns its eventual rejection", async
   const pendingStat = new Promise((_resolve, reject) => {
     rejectLookup = reject;
   });
+  void pendingStat.catch(() => undefined);
   const observed = new Promise<void>((resolve) => {
     started = resolve;
   });
@@ -502,14 +504,20 @@ it("cancels a pending filesystem lookup and owns its eventual rejection", async
     "win32",
     controller.signal,
   );
-  await observed;
-  controller.abort();
-  await expect(pending).rejects.toMatchObject({
-    _tag: "AnalysisCancelledError",
-  });
-  rejectLookup(new Error("filesystem completed after cancellation"));
-  await Promise.resolve();
-  expect(lookup.stat).toHaveBeenCalledTimes(1);
+  try {
+    await awaitAcknowledgement(observed, pending, "apktool filesystem lookup");
+    controller.abort();
+    await expect(pending).rejects.toMatchObject({
+      _tag: "AnalysisCancelledError",
+    });
+    rejectLookup(new Error("filesystem completed after cancellation"));
+    await Promise.resolve();
+    expect(lookup.stat).toHaveBeenCalledTimes(1);
+  } finally {
+    controller.abort();
+    rejectLookup(new Error("fixture cleanup"));
+    await pending.catch(() => undefined);
+  }
 });
 
 it("bounds the entire pending PATH lookup by the selection deadline", async () => {

src/apktool/ApktoolVersionProbe.lifecycle.test.ts

@@ -58,7 +58,6 @@ it.skipIf(process.platform === "win32")(
       await fixture.cleanup([probe]);
     }
   },
-  30_000,
 );
 
 let commands = new ApktoolCommandCleanup();

src/application/AnalysisContextQueries.test.ts

@@ -1,5 +1,6 @@
-import { describe, expect, it } from "vitest";
+import { describe, expect, it, onTestFinished } from "vitest";
 import { createDeferred } from "../../tests/fixtures/binarySession.js";
+import { awaitAcknowledgement } from "../../tests/fixtures/lifecycleAcknowledgement.js";
 
 import {
   getNavigationContext,
@@ -500,7 +501,16 @@ it("waits for late facet cleanup after cancellation and retains its owner", asyn
     settled = true;
     return result;
   });
-  await started.promise;
+  onTestFinished(async () => {
+    controller.abort();
+    late.resolve(err(failure));
+    try {
+      await pending;
+    } finally {
+      await lease.release();
+    }
+  });
+  await awaitAcknowledgement(started.promise, pending, "facet execution");
   controller.abort();
   await Promise.resolve();
   expect(settled).toBe(false);
@@ -530,7 +540,12 @@ it("reports cancellation after a late successful facet that ignored its signal",
     { address: "0x401000", document: "fixture" },
     controller.signal,
   );
-  await started.promise;
+  onTestFinished(async () => {
+    controller.abort();
+    late.resolve(ok(createAnalysisExecution("finished", provider)));
+    await pending;
+  });
+  await awaitAcknowledgement(started.promise, pending, "facet execution");
   controller.abort();
   late.resolve(ok(createAnalysisExecution("finished", provider)));
   const result = await pending;
@@ -586,7 +601,11 @@ it("preserves a provider's cancellation for the same signal", async () => {
     { address: "0x401000", document: "fixture" },
     controller.signal,
   );
-  await started.promise;
+  onTestFinished(async () => {
+    controller.abort(reason);
+    await pending;
+  });
+  await awaitAcknowledgement(started.promise, pending, "provider execution");
   controller.abort(reason);
   const result = await pending;
   if (result.ok) throw new Error("expected cancellation");

src/application/AnalysisProvider.ts

@@ -53,6 +53,11 @@ export interface ExecutionOptions {
    * supplied it.
    */
   readonly selectorOrigin?: "provider";
+  /**
+   * Whether the caller's catalog advertises an operation; refusal recovery
+   * names only advertised operations. Absent means every operation is.
+   */
+  readonly isOperationAdvertised?: (operation: AnalysisOperation) => boolean;
 }
 
 /** Closed operations accepted by provider-backed analysis sessions. */

src/application/AnalysisSettings.test.ts

@@ -6,7 +6,6 @@ import {
   resolveWindowsChangeTimeSettle,
 } from "../config/windowsChangeTimeSettle.js";
 import { pwntoolsInputBytes } from "../native/pwntools/PwntoolsInputLimit.js";
-import { resolveBridgeMaxRequestBytes } from "../config/bridgeMaxRequestBytes.js";
 import { resolveCaptureDecoderHeap } from "../config/captureDecoderHeap.js";
 import { resolveRegexMatchingDeadline } from "../config/regexMatchingDeadline.js";
 import { resolveRegexWorkerHeapRequest } from "../config/regexWorkerHeapRequest.js";
@@ -366,7 +365,6 @@ it("reads an accepted override before an invalid launch seed", () => {
     {
       revision: 1,
       overrides: {
-        "bridge.max_request_bytes": 8192,
         "capture.decoder_old_generation_mib": 64,
         "regex.matching_deadline_ms": 250,
         "regex.worker_heap_request_mib": 128,
@@ -375,9 +373,6 @@ it("reads an accepted override before an invalid launch seed", () => {
       },
     },
     () => ({
-      bridge: resolveBridgeMaxRequestBytes({
-        REA_BRIDGE_MAX_REQUEST_BYTES: "nope",
-      }),
       capture: resolveCaptureDecoderHeap({
         REA_CAPTURE_DECODER_OLD_GENERATION_MIB: "0",
       }),
@@ -393,10 +388,6 @@ it("reads an accepted override before an invalid launch seed", () => {
       }),
     }),
   );
-  expect(resolved.bridge).toMatchObject({
-    maximumBytes: 8192,
-    source: "runtime",
-  });
   expect(resolved.capture).toMatchObject({
     oldGenerationMiB: 64,
     source: "runtime",

src/application/AppBundleExecutable.openStep.test.ts

@@ -4,7 +4,7 @@ import { join } from "node:path";
 import { expect, it } from "vitest";
 
 import { createTestTempDirectory } from "../../tests/fixtures/temporaryDirectory.js";
-import { resolveAppBundleExecutable } from "./AppBundleExecutable.js";
+import { resolveAppleBundleExecutable } from "./AppBundleExecutable.js";
 
 it.each([
   ["no comment", ""],
@@ -22,7 +22,7 @@ it.each([
   await writeFile(infoPlist, source);
   await writeFile(executable, "fixture - never executed");
 
-  expect(await resolveAppBundleExecutable(bundle)).toEqual({
+  expect(await resolveAppleBundleExecutable(bundle)).toEqual({
     ok: true,
     value: { executable, infoPlist },
   });

src/application/AppBundleExecutable.test.ts

@@ -6,7 +6,7 @@ import { describe, expect, it } from "vitest";
 
 import { createTestTempDirectory } from "../../tests/fixtures/temporaryDirectory.js";
 import { binaryPlistObjectTable } from "../../tests/fixtures/binaryPlistObjectTable.js";
-import { resolveAppBundleExecutable } from "./AppBundleExecutable.js";
+import { resolveAppleBundleExecutable } from "./AppBundleExecutable.js";
 
 const binaryPlistExecutableWithThreeByteUid = (): Buffer => {
   const ascii = (value: string): Buffer => {
@@ -53,7 +53,7 @@ it.each([
     );
     await writeFile(executable, "inert fixture; never executed");
 
-    const result = await resolveAppBundleExecutable(bundle);
+    const result = await resolveAppleBundleExecutable(bundle);
     expect(result).toEqual({
       ok: true,
       value: { executable, infoPlist: plistPath },
@@ -71,7 +71,7 @@ describe("app bundle Info.plist magic", () => {
     await writeFile(plistPath, binaryPlistExecutableWithThreeByteUid());
     await writeFile(executable, "inert fixture; never executed");
 
-    expect(await resolveAppBundleExecutable(bundle)).toEqual({
+    expect(await resolveAppleBundleExecutable(bundle)).toEqual({
       ok: true,
       value: { executable, infoPlist: plistPath },
     });
@@ -86,17 +86,13 @@ describe("app bundle Info.plist magic", () => {
       const plist = Buffer.from("bplist00");
       plist[0] = (plist[0] ?? 0) | 0x80;
       await writeFile(plistPath, plist);
-      let decoded = false;
-      const result = await resolveAppBundleExecutable(bundle, {
-        readFile: async () => plist,
-        realpath: async (path) => path,
-        decodeBinaryPlist: async () => {
-          decoded = true;
-          return "Fixture";
+      const result = await resolveAppleBundleExecutable(bundle);
+      expect(result).toMatchObject({
+        ok: false,
+        error: {
+          cause: { name: "XmlPlistTextDecodeError", reason: "invalid-bytes" },
         },
       });
-      expect(decoded).toBe(false);
-      expect(result.ok).toBe(false);
     } finally {
       await rm(root, { recursive: true, force: true });
     }
@@ -118,7 +114,7 @@ describe("app bundle program directory containment", () => {
       await writeFile(join(bundle, "Contents", "Info.plist"), plist);
       let childResolved = false;
 
-      const result = await resolveAppBundleExecutable(bundle, {
+      const result = await resolveAppleBundleExecutable(bundle, {
         readFile: async () => Buffer.from(plist),
         realpath: async (path) => {
           if (path === bundle) return canonicalBundle;

src/application/AppBundleExecutable.ts

@@ -88,10 +88,9 @@ export interface ResolvedAppBundleResourceRoot {
 }
 
 /** Filesystem operations used by the resolver, injectable for boundary tests. */
-export interface AppBundleFileSystem {
+interface AppBundleFileSystem {
   readonly readFile: (path: string) => Promise<Buffer>;
   readonly realpath: (path: string) => Promise<string>;
-  readonly decodeBinaryPlist?: (path: string) => Promise<string>;
 }
 
 const defaultFileSystem: AppBundleFileSystem = { readFile, realpath };
@@ -102,7 +101,7 @@ const defaultFileSystem: AppBundleFileSystem = { readFile, realpath };
  * a wrapper whose `Wrapper` directory holds the iOS bundle; that bundle is
  * found by listing `Wrapper`, never by following the `WrappedBundle` link.
  */
-export const resolveAppBundleExecutable = async (
+const resolveAppBundleExecutable = async (
   path: string,
   fileSystem: AppBundleFileSystem = defaultFileSystem,
 ): Promise<Result<ResolvedAppBundle, BinaryTargetError>> => {
@@ -396,15 +395,6 @@ const codeDescription = (cause: unknown): string => {
     : "";
 };
 
-const isSystemErrorCode = (cause: unknown): boolean => {
-  const code = errorCode(cause);
-  return (
-    typeof code === "string" &&
-    /^[A-Z][A-Z0-9_]+$/u.test(code) &&
-    !code.startsWith("ERR_")
-  );
-};
-
 const isPermissionDenied = (cause: unknown): boolean =>
   errorCode(cause) === "EACCES" || errorCode(cause) === "EPERM";
 
@@ -461,20 +451,12 @@ const resolveLayoutExecutable = async (
 
   let name: string;
   try {
-    name =
-      hasBinaryPlistSignature(plist) &&
-      fileSystem.decodeBinaryPlist !== undefined
-        ? await fileSystem.decodeBinaryPlist(plistPath)
-        : executableName(decodeInfoPlistValue(plist));
+    name = executableName(decodeInfoPlistValue(plist));
   } catch (cause: unknown) {
     return err(
       new BinaryTargetError(
         plistPath,
-        isPermissionDenied(cause)
-          ? `permission denied decoding ${noun} Info.plist ${plistPath}${codeDescription(cause)}`
-          : isSystemErrorCode(cause)
-            ? `could not decode ${noun} Info.plist ${plistPath}${codeDescription(cause)}`
-            : `${noun} Info.plist is malformed, unsupported, or lacks CFBundleExecutable: ${plistPath}${cause instanceof Error ? ` (${cause.message})` : ""}`,
+        `${noun} Info.plist is malformed, unsupported, or lacks CFBundleExecutable: ${plistPath}${cause instanceof Error ? ` (${cause.message})` : ""}`,
         {
           cause,
           // A bundle without code, such as a resource bundle, declares no

src/application/AppBundleOpenStepDepth.test.ts

@@ -3,7 +3,7 @@ import { join } from "node:path";
 
 import { describe, expect, it } from "vitest";
 
-import { resolveAppBundleExecutable } from "./AppBundleExecutable.js";
+import { resolveAppleBundleExecutable } from "./AppBundleExecutable.js";
 import { createTestTempDirectory } from "../../tests/fixtures/temporaryDirectory.js";
 
 describe("OpenStep Info.plist nesting during app target resolution", () => {
@@ -20,7 +20,7 @@ describe("OpenStep Info.plist nesting during app target resolution", () => {
         `{ CFBundleExecutable = App; payload = ${"(".repeat(depth)}x${")".repeat(depth)}; }`,
       );
 
-      const result = await resolveAppBundleExecutable(app);
+      const result = await resolveAppleBundleExecutable(app);
 
       if (depth === 127) {
         expect(result).toMatchObject({
@@ -52,7 +52,7 @@ describe("OpenStep Info.plist nesting during app target resolution", () => {
       `{ CFBundleExecutable = App; }${"(".repeat(3000)}x${")".repeat(3000)}`,
     );
 
-    expect(await resolveAppBundleExecutable(app)).toMatchObject({
+    expect(await resolveAppleBundleExecutable(app)).toMatchObject({
       ok: true,
       value: { executable: join(programs, "App") },
     });

src/application/BinaryTargetResolver.ts

@@ -208,7 +208,7 @@ export const parseBinaryTarget = async (
     throwIfTargetResolutionCancelled(signal);
     const resolved = await resolveAppBundle(canonical);
     throwIfTargetResolutionCancelled(signal);
-    if (!resolved.ok) return err(resolved.error);
+    if (!resolved.ok) return resolved;
     const { executable: path, infoPlist } = resolved.value;
     const bundle =
       infoPlist === undefined ? {} : { bundleInfoPlist: infoPlist };
@@ -747,8 +747,8 @@ const throwIfTargetResolutionCancelled = (signal?: AbortSignal): void => {
 const detectArtifactFormat = async (
   path: string,
   source: StableRegularFileDescriptor,
-  signal?: AbortSignal,
-  sourcePath = path,
+  signal: AbortSignal | undefined,
+  sourcePath: string,
 ): Promise<
   | Exclude<
       BinaryTarget["format"],

src/application/CapabilityInventory.ts

@@ -56,6 +56,7 @@ import {
   type AnalysisCleanupRetry,
 } from "../domain/analysisErrorRecovery.js";
 import { isBundleResourceOperation } from "./binary/BundleResourceSession.js";
+import { offeredOperationAlternatives } from "./OperationAlternatives.js";
 
 const FRIDA_OPERATION_NAMES: ReadonlySet<string> = new Set(
   FRIDA_TOOL_CONTRACTS.map(({ name }) => name),
@@ -126,6 +127,8 @@ export type AvailabilityPolicy = {
 type AvailabilityFacts = {
   readonly defaultModeAvailable?: boolean;
   readonly modes?: readonly ToolAvailabilityMode[];
+  /** Advertised operations the current provider offers for part of the question. */
+  readonly alternatives?: readonly AnalysisOperation[];
 };
 type Availability = AvailabilityFacts &
   (
@@ -162,6 +165,7 @@ type AvailabilityContext = {
   readonly targetFormat: string | undefined;
   readonly descriptors: ReadonlyMap<string, ProviderDescriptor>;
   readonly policy: AvailabilityPolicy;
+  readonly isToolAdvertised: (name: string) => boolean;
 };
 
 type EnhancedContractName = Extract<
@@ -322,10 +326,15 @@ const ADDRESS_CONTEXT_EFFECT_COMPOSITION: ModeComposition = {
   ],
 };
 
-/** Build stable per-operation availability for discovery and tool visibility. */
+/**
+ * Build stable per-operation availability for the caller's advertised tools.
+ * Each tool's availability still derives from the complete provider facts;
+ * advertisement narrows only the reported tools and the alternatives named.
+ */
 export const buildCapabilityInventory = (
   sessionStatus: JsonValue,
   policy: AvailabilityPolicy,
+  isToolAdvertised: (name: string) => boolean = () => true,
 ): readonly ToolAvailability[] => {
   const status = statusSchema.parse(sessionStatus);
   const descriptors = new Map<string, ProviderDescriptor>(
@@ -345,6 +354,7 @@ export const buildCapabilityInventory = (
       targetFormat: status.format,
       descriptors,
       policy,
+      isToolAdvertised,
     });
     const effects = composedEffects(
       contract.name,
@@ -398,8 +408,13 @@ export const buildCapabilityInventory = (
           available: false,
           reason: availability.reason,
           remediation: availability.remediation,
+          ...(availability.alternatives === undefined
+            ? {}
+            : { alternatives: [...availability.alternatives] }),
         };
-  }).sort((left, right) => compareUnicodeCodePoints(left.name, right.name));
+  })
+    .filter(({ name }) => isToolAdvertised(name))
+    .sort((left, right) => compareUnicodeCodePoints(left.name, right.name));
 };
 
 type ClientRequirements = ToolAvailability["client_requirements"];
@@ -445,6 +460,9 @@ export function compactToolAvailability(
         ? {}
         : { default_mode_available: entry.default_mode_available }),
       ...(entry.modes === undefined ? {} : { modes: [...entry.modes] }),
+      ...(entry.available || entry.alternatives === undefined
+        ? {}
+        : { alternatives: [...entry.alternatives] }),
       ...(JSON.stringify(entry.client_requirements) ===
       JSON.stringify(clientRequirements)
         ? {}
@@ -959,6 +977,7 @@ const providerAvailability = ({
   kind,
   name,
   policy,
+  isToolAdvertised,
 }: AvailabilityContext): Availability => {
   if (kind === "enhanced") return composedAvailability(name, descriptors);
   const launchFailure =
@@ -976,12 +995,20 @@ const providerAvailability = ({
       kind === "native-provider")
   )
     return { reason: "available", remediation: null };
-  if (descriptor === undefined)
+  if (descriptor === undefined) {
+    const offered = offeredOperationAlternatives(
+      name,
+      (alternative) =>
+        descriptors.get(alternative)?.available === true &&
+        isToolAdvertised(alternative),
+    );
     return {
       reason: "provider_missing",
       remediation:
         "Install or configure a provider that declares this operation.",
+      ...(offered === undefined ? {} : { alternatives: offered.operations }),
     };
+  }
   return descriptor.available
     ? { reason: "available", remediation: null }
     : {

src/application/CapabilityScope.test.ts

@@ -93,7 +93,12 @@ it("joins concurrent shutdowns and seals ownership before asynchronous cleanup s
   try {
     await awaitAcknowledgement(entered, first, "capability cleanup");
     second = scope.close();
-    expect(second).toBe(first);
+    let secondSettled = false;
+    void second.then(() => {
+      secondSettled = true;
+    });
+    await Promise.resolve();
+    expect(secondSettled).toBe(false);
     expect(close).toHaveBeenCalledTimes(1);
     expect(() => scope.own("later-owner", {}, async () => undefined)).toThrow(
       "after host shutdown",

src/application/ClaudeDesktopConfigLocation.test.ts

@@ -1,15 +1,14 @@
+import { CLAUDE_DESKTOP_PACKAGE_FAMILY } from "../../tests/fixtures/claudeDesktop.js";
 import { basename, dirname, join, sep } from "node:path";
 
 import { describe, expect, it } from "vitest";
 
 import {
-  CLAUDE_DESKTOP_PACKAGE_FAMILY,
   resolveWindowsClaudeDesktopLocation,
   type ClaudeDesktopLocationFileSystem,
 } from "./ClaudeDesktopConfigLocation.js";
 import {
   packagePrivateLauncherBlocker,
-  packagePrivateStorageFamily,
   redirectedWriteBlocker,
   type MsixPackageView,
   windowsAppDataRoots,
@@ -642,15 +641,19 @@ describe("package-private REA launchers", () => {
     "rea.mjs",
   );
 
-  it("names the package that privately stores a path", () => {
-    expect(packagePrivateStorageFamily(stored, packages)).toBe(
-      CLAUDE_DESKTOP_PACKAGE_FAMILY,
-    );
-    expect(packagePrivateStorageFamily(launcher, packages)).toBeUndefined();
+  it("blocks private cache paths without treating other package paths as private", () => {
+    const identity = (path: string) => path;
     expect(
-      packagePrivateStorageFamily(
-        join(packages, CLAUDE_DESKTOP_PACKAGE_FAMILY, "LocalState", "x"),
+      packagePrivateLauncherBlocker([stored], packages, identity),
+    ).toContain(CLAUDE_DESKTOP_PACKAGE_FAMILY);
+    expect(
+      packagePrivateLauncherBlocker([launcher], packages, identity),
+    ).toBeUndefined();
+    expect(
+      packagePrivateLauncherBlocker(
+        [join(packages, CLAUDE_DESKTOP_PACKAGE_FAMILY, "LocalState", "x")],
         packages,
+        identity,
       ),
     ).toBeUndefined();
   });

src/application/ClaudeDesktopConfigLocation.ts

@@ -29,7 +29,7 @@ const PROBES_PER_DIRECTORY = 2;
  * publisher ID Windows derives from `CN="Anthropic, PBC", O="Anthropic, PBC",
  * L=San Francisco, S=California, C=US, …`.
  */
-export const CLAUDE_DESKTOP_PACKAGE_FAMILY = "Claude_pzs8sxrjxfjjc";
+const CLAUDE_DESKTOP_PACKAGE_FAMILY = "Claude_pzs8sxrjxfjjc";
 
 /** Observed Claude package names plus Windows' 13-character publisher ID. */
 const CLAUDE_PACKAGE_FAMILY =

src/application/ClientConfigurationFiles.ts

@@ -23,14 +23,14 @@ export interface ClientConfigurationSource extends ClientConfigurationDocument {
 const invalidEnvironment = (
   path: string,
   error: ConfigurationError,
-  overrides: Readonly<Record<string, string>> = {},
-  platform: NodeJS.Platform = process.platform,
+  overrides: Readonly<Record<string, string>> | undefined,
+  platform: NodeJS.Platform,
 ): ClientConfigurationFileError => {
   const constraints = error.settings
     .map(({ setting, constraint }) => `${setting}: ${constraint}`)
     .join("; ");
   const overrideNames = new Set(
-    Object.keys(overrides).map((name) =>
+    Object.keys(overrides ?? {}).map((name) =>
       platform === "win32" ? name.toUpperCase() : name,
     ),
   );

src/application/ClientRegistrationEnvironment.ts

@@ -1,15 +1,9 @@
+import { validateRegistrationEnvironment } from "./RegistrationEnvironmentValidation.js";
 import { clientRegistrationSource } from "./ClientRegistrationSource.js";
 import type { SetupClient } from "./SupportedClients.js";
 
-import {
-  inspectAnalysisSettingSeeds,
-  launchAnalysisSetting,
-} from "../config/analysisSettings.js";
-import {
-  parseEnvironment,
-  parseEnvironmentValues,
-  type Environment,
-} from "../config/environment.js";
+import { launchAnalysisSetting } from "../config/analysisSettings.js";
+import type { Environment } from "../config/environment.js";
 import {
   DEFAULT_GHIDRA_STARTUP_TIMEOUT_MS,
   DEFAULT_HOPPER_STARTUP_TIMEOUT_MS,
@@ -314,34 +308,24 @@ export const clientRegistrationEnvironmentDetailsWithBudget = async (
 const validatedDetails = (
   environment: Readonly<Record<string, string>>,
   validation: "fragment" | "effective",
-  platform: NodeJS.Platform = process.platform,
-  expressionSettings: ReadonlySet<string> = new Set(),
+  platform: NodeJS.Platform,
+  expressionSettings: ReadonlySet<string>,
 ): Result<ClientRegistrationEnvironmentDetails, ConfigurationError> => {
-  const runtimeEnvironment = snapshotEnvironment(environment, platform);
-  const unresolved = new Set(
-    Object.keys(
-      snapshotEnvironment(
-        Object.fromEntries([...expressionSettings].map((name) => [name, ""])),
-        platform,
-      ),
-    ),
+  const validated = validateRegistrationEnvironment(
+    environment,
+    expressionSettings,
+    validation,
+    platform,
   );
-  const validated =
-    validation === "fragment"
-      ? parseEnvironmentValues(runtimeEnvironment, unresolved)
-      : parseEnvironment(runtimeEnvironment, unresolved);
-  if (!validated.ok) return validated;
-  const seedIssues = inspectAnalysisSettingSeeds(
-    runtimeEnvironment,
-    unresolved,
-  );
-  if (seedIssues.length > 0)
-    return err(
-      new ConfigurationError("Invalid REA environment configuration", {
-        settings: seedIssues,
-      }),
-    );
-  return ok(environmentDetails(environment, validated.value, unresolved));
+  return validated.ok
+    ? ok(
+        environmentDetails(
+          environment,
+          validated.value.configuration,
+          validated.value.unresolvedSettings,
+        ),
+      )
+    : validated;
 };
 
 const admittedTimeout = (
@@ -397,7 +381,7 @@ const registrationTimeoutInputs = (
 const environmentDetails = (
   environment: Readonly<Record<string, string>>,
   configuration: Environment,
-  unresolvedSettings: ReadonlySet<string> = new Set(),
+  unresolvedSettings: ReadonlySet<string>,
 ): ClientRegistrationEnvironmentDetails => ({
   values: Object.fromEntries(
     Object.entries(environment).sort(([left], [right]) =>

src/application/ClientRegistrationStatus.ts

@@ -649,7 +649,7 @@ const packageVersionRemediation = (
 const unavailableStatus = (
   client: SetupClient,
   state: "missing" | "invalid",
-  configPath: string = client.configPath,
+  configPath: string,
 ): ClientRegistrationStatus => ({
   client: client.name,
   config_path: configPath,

src/application/DirectAnalysis.ts

@@ -67,10 +67,7 @@ import type { AnalysisProfileCommitment } from "../domain/analysisProfile.js";
 import { profilelessNativeDispatchProfile } from "./binary/ProfilelessNativeWorkflowSnapshot.js";
 import { err, ok, type Result } from "../domain/result.js";
 import type { AnalysisSnapshot } from "../domain/analysisSnapshot.js";
-import type {
-  AnalysisExecution,
-  ProviderIdentity,
-} from "./AnalysisProvider.js";
+import type { AnalysisExecution } from "./AnalysisProvider.js";
 import { createAnalysisExecutionEvidence } from "../domain/analysisObservation.js";
 import { REA_WORKFLOW_PROVIDER } from "../domain/investigationProviders.js";
 import { workflowAnalysisProfile } from "./WorkflowAnalysisProfile.js";
@@ -296,7 +293,7 @@ export const runProviderAnalysis = async (
               delivery.detail,
               operationSignal,
             );
-      return result.ok ? ok(result.value) : err(cliError(result.error, path));
+      return result.ok ? result : err(cliError(result.error, path));
     }
     const admitted = admitOneShotOperation(tool, arguments_, operationSignal);
     if (!admitted.ok) return err(cliError(admitted.error, path));
@@ -445,14 +442,14 @@ const runAnalysis = async (
     let isDirectory = false;
     if (BUNDLE_RESOURCE_OPERATIONS.has(tool)) {
       const inspected = await inspectBundleResourceInput(path);
-      if (!inspected.ok) return err(inspected.error);
+      if (!inspected.ok) return inspected;
       isDirectory = inspected.value;
     }
     if (BUNDLE_RESOURCE_OPERATIONS.has(tool) && isDirectory) {
       const opened = await session.openBundleResources(path, { signal });
-      if (!opened.ok) return err(opened.error);
+      if (!opened.ok) return opened;
       const execution = await session.execute(tool, arguments_, { signal });
-      if (!execution.ok) return err(execution.error);
+      if (!execution.ok) return execution;
       return ok(
         withEvidenceCallerInputPath(
           createAnalysisExecutionEvidence(
@@ -472,7 +469,7 @@ const runAnalysis = async (
       signal,
       ...targetSelection(options),
     });
-    if (!prepared.ok) return err(prepared.error);
+    if (!prepared.ok) return prepared;
     const { snapshot } = prepared.value;
     let resolvedTarget: ResolvedSessionOpen | undefined;
     if (snapshot !== undefined && prepared.value.target !== undefined) {
@@ -484,7 +481,7 @@ const runAnalysis = async (
           ? {}
           : { providerId: options.providerId }),
       });
-      if (!preview.ok) return err(preview.error);
+      if (!preview.ok) return preview;
       resolvedTarget = preview.value;
       const route = preview.value.route;
       const bindingProfile =
@@ -548,7 +545,7 @@ const runAnalysis = async (
       resolvedTarget === undefined
         ? await session.open(path, openOptions)
         : await session.openResolvedTarget(resolvedTarget, openOptions);
-    if (!opened.ok) return err(opened.error);
+    if (!opened.ok) return opened;
     const evidenceProfile = analysisProfileForEvidence(session, tool);
     const executed = await executeAnalysisTool({
       session,

src/application/DshCordisPatch.ts

@@ -1,18 +1,11 @@
-import { inspectAnalysisSettingSeeds } from "../config/analysisSettings.js";
+import { validateRegistrationEnvironment } from "./RegistrationEnvironmentValidation.js";
 import { isMap, isSeq, type Node, type YAMLMap } from "yaml";
-import {
-  parseEnvironment,
-  parseEnvironmentValues,
-  type Environment,
-} from "../config/environment.js";
+import type { Environment } from "../config/environment.js";
 import { ConfigurationError } from "../domain/configurationErrors.js";
 import { err, ok, type Result } from "../domain/result.js";
 import type { ClientConfigurationDocument } from "./ClientConfigurationDocument.js";
 import { CordisExpression, DshCordisYaml } from "./DshCordisYaml.js";
-import {
-  mergeEnvironmentOverrides,
-  snapshotEnvironment,
-} from "../process/snapshotEnvironment.js";
+import { mergeEnvironmentOverrides } from "../process/snapshotEnvironment.js";
 
 export const DSH_ENTRY_ID = "rea-mcp";
 export const DSH_CLIENT_PACKAGE = "@deepseek-ai/dsh-mcp-client";
@@ -255,35 +248,15 @@ export class DshCordisDocument {
         );
       literals[key] = value;
     }
-    const runtimeEnvironment = snapshotEnvironment(literals, platform);
-    const unresolved = new Set(
-      Object.keys(
-        snapshotEnvironment(
-          Object.fromEntries([...unresolvedSettings].map((key) => [key, ""])),
-          platform,
-        ),
-      ),
+    const validated = validateRegistrationEnvironment(
+      literals,
+      unresolvedSettings,
+      validation,
+      platform,
     );
-    const validated =
-      validation === "fragment"
-        ? parseEnvironmentValues(runtimeEnvironment, unresolved)
-        : parseEnvironment(runtimeEnvironment, unresolved);
-    if (!validated.ok) return validated;
-    const seedIssues = inspectAnalysisSettingSeeds(
-      runtimeEnvironment,
-      unresolved,
-    );
-    if (seedIssues.length > 0)
-      return err(
-        new ConfigurationError("Invalid REA environment configuration", {
-          settings: seedIssues,
-        }),
-      );
-    return ok({
-      values: literals,
-      configuration: validated.value,
-      unresolvedSettings: unresolved,
-    });
+    return validated.ok
+      ? ok({ values: literals, ...validated.value })
+      : validated;
   }
 
   desired(

src/application/EvidenceBundleCommands.ts

@@ -307,9 +307,7 @@ const projectWrite = (
   bundle: EvidenceBundle,
   written: Result<PublishedEvidenceFile, AnalysisError>,
 ): Result<JsonValue, AnalysisError> =>
-  written.ok
-    ? ok(evidenceBundleExportReceipt(bundle, written.value))
-    : err(written.error);
+  written.ok ? ok(evidenceBundleExportReceipt(bundle, written.value)) : written;
 
 /** Standalone bundle discovery shares the same read owner as import and comparison. */
 export const inspectEvidenceBundleCommand = async (

src/application/EvidenceBundleFiles.ts

@@ -31,16 +31,11 @@ import {
 import { readJsonFile, writeTextParts } from "./JsonFiles.js";
 import type { ArtifactResourceScope } from "../artifacts/ArtifactResourceScope.js";
 
-type EvidenceReadFailure =
+type EvidenceBundleFileFailure =
   | EvidenceFileError
   | EvidenceIntegrityError
   | AnalysisResourceConstraintError
   | AnalysisCancelledError;
-type EvidenceWriteFailure =
-  | EvidenceFileError
-  | EvidenceIntegrityError
-  | AnalysisCancelledError
-  | AnalysisResourceConstraintError;
 
 /** Project a published bundle with record and residual-unknown revision counts. */
 export const evidenceBundleExportReceipt = (
@@ -59,7 +54,7 @@ export const readEvidenceBundle = (
   signal?: AbortSignal,
   resources?: ArtifactResourceScope,
   operation = "import_evidence_bundle",
-): Promise<Result<EvidenceBundle, EvidenceReadFailure>> =>
+): Promise<Result<EvidenceBundle, EvidenceBundleFileFailure>> =>
   readBundle(path, false, signal, resources, operation);
 
 /** Read an authenticated immutable snapshot for a subsequent session import. */
@@ -68,7 +63,7 @@ export const readImmutableEvidenceBundle = (
   signal?: AbortSignal,
   resources?: ArtifactResourceScope,
   operation = "import_evidence_bundle",
-): Promise<Result<EvidenceBundle, EvidenceReadFailure>> =>
+): Promise<Result<EvidenceBundle, EvidenceBundleFileFailure>> =>
   readBundle(path, true, signal, resources, operation);
 
 /** Join both comparison reads and preserve cleanup owned by either failed input. */
@@ -109,7 +104,7 @@ const readBundle = async (
   signal?: AbortSignal,
   resources?: ArtifactResourceScope,
   operation = "import_evidence_bundle",
-): Promise<Result<EvidenceBundle, EvidenceReadFailure>> => {
+): Promise<Result<EvidenceBundle, EvidenceBundleFileFailure>> => {
   if (signal?.aborted)
     return err(
       new AnalysisCancelledError(operation, {
@@ -135,7 +130,7 @@ export const authenticateEvidenceBundleValue = (
   value: unknown,
   path: string,
   options: { readonly operation: string; readonly immutable?: boolean },
-): Result<EvidenceBundle, EvidenceReadFailure> => {
+): Result<EvidenceBundle, EvidenceBundleFileFailure> => {
   const { operation } = options;
   const headroom = requireAuthenticationHeadroom(value, {
     operation,
@@ -226,7 +221,7 @@ export const writeEvidenceBundle = async (
 ): Promise<
   Result<
     { readonly path: string; readonly bytes: number },
-    EvidenceWriteFailure
+    EvidenceBundleFileFailure
   >
 > => {
   if (signal?.aborted === true)

src/application/GeminiClientSettings.ts

@@ -2,6 +2,7 @@ import { realpath, stat } from "node:fs/promises";
 import { dirname, resolve } from "node:path";
 import { stripComments } from "jsonc-parser";
 import { z } from "zod";
+import { jsonRecordSchema } from "../domain/jsonRecord.js";
 import { err, ok, type Result } from "../domain/result.js";
 import { PRODUCT_IDENTITY } from "../identity.js";
 import type { ArtifactResourceScope } from "../artifacts/ArtifactResourceScope.js";
@@ -34,8 +35,7 @@ const folderTrustSchema = z
       .optional(),
   })
   .passthrough();
-const trustRulesSchema = z.record(
-  z.string(),
+const trustRulesSchema = jsonRecordSchema(
   z.enum(["TRUST_FOLDER", "TRUST_PARENT", "DO_NOT_TRUST"]),
 );
 const sourceFailure = (
@@ -47,18 +47,18 @@ const sourceFailure = (
   detail: `Gemini settings in ${path} are malformed: ${cause instanceof Error ? cause.message : String(cause)}`,
 });
 
-// Gemini resolves symlinks when possible and compares macOS/Windows paths without case.
-const normalizedPath = async (
-  path: string,
-  platform: NodeJS.Platform,
-): Promise<string> => {
-  const physical = await realpath(path).catch(() => path);
-  const absolute = resolve(physical).replaceAll("\\", "/");
+// Gemini normalizes trust-file keys before replacement and rule precedence.
+const normalizedPath = (path: string, platform: NodeJS.Platform): string => {
+  const absolute = resolve(path).replaceAll("\\", "/");
   return platform === "win32" || platform === "darwin"
     ? absolute.toLowerCase()
     : absolute;
 };
 
+// Applicability additionally resolves symlinks, without changing rule identity.
+const physicalPath = async (path: string, platform: NodeJS.Platform) =>
+  normalizedPath(await realpath(path).catch(() => path), platform);
+
 const workspaceTrusted = async (
   settings: NonNullable<SetupClient["geminiSettings"]>,
   sources: readonly SettingsSource[],
@@ -78,21 +78,29 @@ const workspaceTrusted = async (
   );
   if (!text.ok) return text;
   if (text.value === undefined) return ok(false);
-  let rules: z.output<typeof trustRulesSchema>;
+  let rules: ReadonlyMap<string, z.output<typeof trustRulesSchema>[string]>;
   try {
     // Gemini strips comments, but rejects trailing commas, BOMs and invalid trust levels.
-    rules = trustRulesSchema.parse(JSON.parse(stripComments(text.value)));
+    const parsed = trustRulesSchema.parse(
+      JSON.parse(stripComments(text.value)),
+    );
+    rules = new Map(
+      Object.entries(parsed).map(([path, level]) => [
+        normalizedPath(path, settings.platform),
+        level,
+      ]),
+    );
   } catch (cause: unknown) {
     return err(sourceFailure(settings.trustedFoldersPath, cause));
   }
-  const workspace = await normalizedPath(
+  const workspace = await physicalPath(
     settings.workspaceDirectory,
     settings.platform,
   );
   let longest = -1;
   let trusted = false;
-  for (const [path, level] of Object.entries(rules)) {
-    const effective = await normalizedPath(
+  for (const [path, level] of rules) {
+    const effective = await physicalPath(
       level === "TRUST_PARENT" ? dirname(path) : path,
       settings.platform,
     );
@@ -138,11 +146,11 @@ export const readClientPolicyBlock = async (
     if (parsed.value !== undefined)
       sources.push({ path, document: parsed.value });
   }
-  const workspace = await normalizedPath(
+  const workspace = await physicalPath(
     settings.workspaceDirectory,
     settings.platform,
   );
-  const home = await normalizedPath(
+  const home = await physicalPath(
     dirname(dirname(client.configPath)),
     settings.platform,
   );

src/application/GooseConfigurationYaml.test.ts

@@ -4,7 +4,6 @@ import { parseDocument } from "yaml";
 import { ConfigurationError } from "../domain/configurationErrors.js";
 import {
   editGooseConfigurationYaml,
-  GooseConfigurationYamlError,
   parseGooseConfigurationYaml,
   stringifyGooseConfigurationYaml,
 } from "./GooseConfigurationYaml.js";
@@ -111,11 +110,15 @@ describe("Goose YAML 1.2 source facts", () => {
       } catch (cause: unknown) {
         failure = cause;
       }
-      expect(failure).toBeInstanceOf(GooseConfigurationYamlError);
+      expect(failure).toBeInstanceOf(ConfigurationError);
       expect(failure).toMatchObject({ kind: "shadowed_member", path });
+      expect(failure).toHaveProperty(
+        "message",
+        `Goose YAML contains a duplicate mapping member at ${path.join(".")}`,
+      );
       expect(() =>
         editGooseConfigurationYaml(source, ["extensions", "rea"], extension),
-      ).toThrow(GooseConfigurationYamlError);
+      ).toThrow(`duplicate mapping member at ${path.join(".")}`);
     },
   );
 

src/application/GooseConfigurationYaml.ts

@@ -143,7 +143,7 @@ const gooseYamlOptions = {
 } as const;
 
 /** A duplicate YAML member cannot establish one selected configuration value. */
-export class GooseConfigurationYamlError extends ConfigurationError {
+class GooseConfigurationYamlError extends ConfigurationError {
   readonly kind = "shadowed_member";
   constructor(
     readonly path: readonly string[],

src/application/GuidanceAccess.test.ts

@@ -1,7 +1,6 @@
 import { expect, it } from "vitest";
 import { ArtifactResourceScope } from "../artifacts/ArtifactResourceScope.js";
 import { RegularFileCleanupFailure } from "../filesystem/OwnedRegularFile.js";
-import { ArtifactReaderFailure } from "../artifacts/ArtifactReader.js";
 import { MANAGED_SKILL_FILES } from "./SetupSkill.js";
 import { searchGuidanceOutputSchema } from "../contracts/guidanceContracts.js";
 import { RetainedRegularFileCleanupFailure } from "./RegularFileRead.js";

src/application/HopperDownload.ts

@@ -49,13 +49,13 @@ export type HopperRequestFailure = HopperDownloadFailure & {
 export const HOPPER_DOWNLOAD_BYTES = 100_000_000;
 
 /** Longest wait for the response or its next bytes before a download stalls. */
-export const HOPPER_DOWNLOAD_STALL_MS = 30_000;
+const HOPPER_DOWNLOAD_STALL_MS = 30_000;
 
 /**
  * Overall bound for one download. A slow connection that keeps delivering
  * bytes is bounded here rather than by a deadline sized for fast networks.
  */
-export const HOPPER_DOWNLOAD_TIMEOUT_MS = 15 * 60_000;
+const HOPPER_DOWNLOAD_TIMEOUT_MS = 15 * 60_000;
 
 /** Small release feed and key requests must not use the package bounds. */
 export const HOPPER_METADATA_TIMEOUT_MS = 10_000;

src/application/LinuxHopper.test.ts

@@ -4,9 +4,8 @@ import { join } from "node:path";
 import { describe, expect, it, vi } from "vitest";
 
 import {
-  formatCommandForDisplay,
   installLinuxHopper,
-  linuxPackageManagerCommands,
+  linuxHopperInstallDisclosure,
   linuxSharedLibrariesAvailable,
   parseLinuxDistribution,
   supportsReaHost,
@@ -31,19 +30,6 @@ import * as nativeCommands from "../process/ExecFileOutput.js";
 import * as pythonRuntime from "./LinuxHopperPython.js";
 import { createTestTempDirectory } from "../../tests/fixtures/temporaryDirectory.js";
 
-describe("Linux package-manager command display", () => {
-  it("preserves argument boundaries for paths with spaces and quotes", () => {
-    expect(
-      formatCommandForDisplay("pkexec", [
-        "pacman",
-        "-U",
-        "/tmp/rea test/hopper.arch",
-        "apostrophe'path",
-      ]),
-    ).toBe('pkexec pacman -U "/tmp/rea test/hopper.arch" "apostrophe\'path"');
-  });
-});
-
 class RecordingLinuxHost implements LinuxHopperInstallHost {
   distributionValue: LinuxDistribution | undefined = {
     id: "ubuntu",
@@ -263,47 +249,34 @@ describe("Linux Hopper installation", () => {
   ] as const)(
     "selects the %s native package manager and dependencies",
     (family, executable, packages) => {
-      expect(
-        linuxPackageManagerCommands(family, "/tmp/Hopper package", true),
-      ).toContainEqual(
-        expect.objectContaining({
-          executable,
-          args: expect.arrayContaining([...packages]),
-        }),
+      const root = linuxHopperInstallDisclosure(family, true).commands;
+      const nonRoot = linuxHopperInstallDisclosure(family, false).commands;
+      expect(root.some((command) => command.startsWith(`${executable} `))).toBe(
+        true,
       );
+      for (const dependency of packages)
+        expect(
+          root.some((command) => command.split(" ").includes(dependency)),
+        ).toBe(true);
       expect(
-        linuxPackageManagerCommands(family, "/tmp/Hopper package", false),
-      ).toContainEqual(
-        expect.objectContaining({
-          executable: "pkexec",
-          args: expect.arrayContaining([executable, "/tmp/Hopper package"]),
-        }),
-      );
+        nonRoot.some(
+          (command) =>
+            command.startsWith(`pkexec ${executable} `) &&
+            command.includes(`"<verified-hopper.${family}>"`),
+        ),
+      ).toBe(true);
     },
   );
 
   it.each([true, false])(
     "keeps installed apt dependencies at their versions while Hopper may upgrade (root: %s)",
     (isRoot) => {
-      const prefix = isRoot ? [] : ["pkexec"];
+      const prefix = isRoot ? "" : "pkexec ";
       expect(
-        linuxPackageManagerCommands("deb", "/tmp/Hopper package", isRoot).map(
-          ({ executable, args }) => [executable, ...args],
-        ),
+        linuxHopperInstallDisclosure("deb", isRoot).commands.slice(0, 2),
       ).toEqual([
-        [
-          ...prefix,
-          "apt-get",
-          "install",
-          "-y",
-          "--no-upgrade",
-          "xvfb",
-          "xauth",
-          "python3",
-          "libx11-6",
-          "libxtst6",
-        ],
-        [...prefix, "apt-get", "install", "-y", "/tmp/Hopper package"],
+        `${prefix}apt-get install -y --no-upgrade xvfb xauth python3 libx11-6 libxtst6`,
+        `${prefix}apt-get install -y "<verified-hopper.deb>"`,
       ]);
     },
   );

src/application/LinuxHopper.ts

@@ -511,7 +511,7 @@ export const installLinuxHopper = async (
 };
 
 /** Canonical launcher path for a legacy user-local Linux Hopper installation. */
-export const linuxHopperLauncherPath = (home: string): string =>
+const linuxHopperLauncherPath = (home: string): string =>
   appendCallerPath(home, ".local/share/rea/hopper/bin/Hopper", "linux");
 
 /** REA's managed Linux install path, preferred when no explicit launcher is set. */
@@ -796,7 +796,7 @@ const runPackageManagerCommand = async (
 };
 
 /** Preserve argument boundaries in a command shown to the user. */
-export const formatCommandForDisplay = (
+const formatCommandForDisplay = (
   executable: string,
   args: readonly string[],
 ): string =>
@@ -807,7 +807,7 @@ export const formatCommandForDisplay = (
     .join(" ");
 
 /** Select the native package manager and authorization boundary without shell evaluation. */
-export const linuxPackageManagerCommands = (
+const linuxPackageManagerCommands = (
   family: LinuxPackageFamily,
   archive: string,
   isRoot: boolean,

src/application/LinuxHopperPackageRuntime.test.ts

@@ -6,16 +6,11 @@ import { err, ok } from "../domain/result.js";
 import * as pythonRuntime from "./LinuxHopperPython.js";
 import {
   linuxHopperInstallDisclosure,
-  linuxPackageManagerCommands,
   systemLinuxHopperInstallHost,
 } from "./LinuxHopper.js";
 
 /** Package transactions that install Hopper before any runtime library repair. */
-const debCommands = linuxPackageManagerCommands(
-  "deb",
-  "/tmp/verified-hopper.deb",
-  false,
-).length;
+const debCommands = 2;
 
 const missingPython = {
   interpreter: "/usr/bin/python3",
@@ -41,6 +36,34 @@ const commandSuccess = {
 };
 
 describe("Linux Hopper package runtime", () => {
+  it("preserves package path boundaries in the reported failed command", async () => {
+    const directory = await createTestTempDirectory("rea-hopper-command-");
+    const host = systemLinuxHopperInstallHost(
+      { XDG_CACHE_HOME: directory },
+      async () => undefined,
+    );
+    const archive = "/tmp/rea test/hopper's package.rpm";
+    const command = vi
+      .spyOn(boundedCommands, "execFileBoundedOutput")
+      .mockResolvedValue({
+        exitCode: 1,
+        stdout: "",
+        stderr: "fixture failure",
+        outputTruncated: false,
+      });
+    try {
+      const result = await host.installPackage("rpm", archive);
+      expect(command.mock.calls[0]?.[1]).toContain(archive);
+      expect(result).toMatchObject({
+        status: "failed",
+        failure: { command: expect.stringContaining(`"${archive}"`) },
+      });
+    } finally {
+      command.mockRestore();
+      await host.close?.();
+    }
+  });
+
   it.each([
     {
       name: "installs and verifies the matching future Python library",

src/application/MacHopperPaths.test.ts

@@ -11,7 +11,7 @@ import {
 import { tmpdir } from "node:os";
 import { join, posix, relative } from "node:path";
 import { installMacHopperBundle } from "./MacHopper.js";
-import { linuxHopperLauncherPath } from "./LinuxHopper.js";
+import { linuxHopperLauncherCandidates } from "./LinuxHopper.js";
 import { homeDirectoryFromEnvironment } from "../config/homeDirectory.js";
 import { describe, expect, it } from "vitest";
 
@@ -19,34 +19,17 @@ import {
   localMacHopperRenamedLauncherCandidates,
   macHopperCanonicalLauncherCandidates,
   macHopperUserApplicationPath,
-  macHopperRenamedLauncherCandidates,
   readMacHopperRenamedLauncherCandidates,
 } from "./MacHopperPaths.js";
 
 describe("macOS Hopper launcher paths", () => {
   const home = "/fixture/home";
 
-  it("keeps canonical system and user candidates ahead of renamed bundles", () => {
+  it("keeps canonical system and user candidates in order", () => {
     expect(macHopperCanonicalLauncherCandidates(home)).toEqual([
       "/Applications/Hopper Disassembler.app/Contents/MacOS/hopper",
       "/fixture/home/Applications/Hopper Disassembler.app/Contents/MacOS/hopper",
     ]);
-    expect(
-      macHopperRenamedLauncherCandidates(
-        home,
-        new Map([
-          [
-            "/Applications",
-            ["Hopper z.app", "Hopper a.app", "Hopper Disassembler.app"],
-          ],
-          ["/fixture/home/Applications", ["Hopper User.app"]],
-        ]),
-      ),
-    ).toEqual([
-      "/Applications/Hopper a.app/Contents/MacOS/hopper",
-      "/Applications/Hopper z.app/Contents/MacOS/hopper",
-      "/fixture/home/Applications/Hopper User.app/Contents/MacOS/hopper",
-    ]);
   });
 
   it.skipIf(process.platform === "win32")(
@@ -84,7 +67,7 @@ describe("macOS Hopper launcher paths", () => {
         expect(macHopperCanonicalLauncherCandidates(home)[1]).toBe(
           `${home}/Applications/Hopper Disassembler.app/Contents/MacOS/hopper`,
         );
-        expect(linuxHopperLauncherPath(home)).toBe(
+        expect(linuxHopperLauncherCandidates(home)[1]).toBe(
           `${home}/.local/share/rea/hopper/bin/Hopper`,
         );
         const source = join(temporary, "source.app");
@@ -154,13 +137,17 @@ describe("macOS Hopper launcher paths", () => {
         recursive: true,
       });
       await mkdir(join(applications, "Hopper Disassembler.app"));
+      await mkdir(join(applications, "Hopper Alpha.app"));
+      await mkdir(join(applications, "Hopper Zeta.app"));
       await writeFile(join(applications, "Hopper File.app"), "");
       const runtime = [...localMacHopperRenamedLauncherCandidates(home)];
       expect(runtime.filter((path) => path.startsWith(home))).toEqual([
+        posix.join(home, "Applications/Hopper Alpha.app/Contents/MacOS/hopper"),
         posix.join(
           home,
           "Applications/Hopper Preview.app/Contents/MacOS/hopper",
         ),
+        posix.join(home, "Applications/Hopper Zeta.app/Contents/MacOS/hopper"),
       ]);
       expect(await readMacHopperRenamedLauncherCandidates(home)).toEqual(
         runtime,

src/application/MacHopperPaths.ts

@@ -43,7 +43,7 @@ export const macHopperCanonicalLauncherCandidates = (
 ];
 
 /** Build deterministic candidates from the discovered app bundle names. */
-export const macHopperRenamedLauncherCandidates = (
+const macHopperRenamedLauncherCandidates = (
   homeDirectory: string,
   bundleNamesByRoot: ReadonlyMap<string, readonly string[]>,
 ): readonly string[] =>

src/application/McpClientRegistration.test.ts

@@ -12,6 +12,7 @@ import { join, sep } from "node:path";
 import { afterEach, beforeEach, describe, expect, it, vi } from "vitest";
 
 import { createTestTempDirectory } from "../../tests/fixtures/temporaryDirectory.js";
+import { awaitAcknowledgement } from "../../tests/fixtures/lifecycleAcknowledgement.js";
 import { ok } from "../domain/result.js";
 import { ArtifactResourceScope } from "../artifacts/ArtifactResourceScope.js";
 let resources: ArtifactResourceScope;
@@ -612,17 +613,22 @@ describe("Registration cancellation and sibling outcomes", () => {
       entered();
       await blocked;
     });
-    await admitted;
-    const controller = new AbortController();
-    const queued = register(path, launcher, controller.signal);
-    controller.abort("queued cancellation");
-    unblock();
-    await first;
-    expect(await queued).toMatchObject({
-      ok: false,
-      error: { _tag: "AnalysisCancelledError" },
-    });
-    expect(await readFile(path, "utf8")).toBe("{}");
+    try {
+      await awaitAcknowledgement(admitted, first, "registration transaction");
+      const controller = new AbortController();
+      const queued = register(path, launcher, controller.signal);
+      controller.abort("queued cancellation");
+      unblock();
+      await first;
+      expect(await queued).toMatchObject({
+        ok: false,
+        error: { _tag: "AnalysisCancelledError" },
+      });
+      expect(await readFile(path, "utf8")).toBe("{}");
+    } finally {
+      unblock();
+      await first;
+    }
   });
 
   it("keeps successful siblings after ordinary failures and marks work not started after cancellation", async () => {

src/application/McpRegistrationCatalog.test.ts

@@ -5,7 +5,28 @@ import { z } from "zod";
 
 import { createTestTempDirectory } from "../../tests/fixtures/temporaryDirectory.js";
 import { createTestProcessRegistry } from "../../tests/support/process/processFixture.js";
-import { resolveMcpRegistrationClientId } from "./McpRegistrationCatalog.js";
+import { resolveMcpRegistrationClients } from "./McpRegistrationCatalog.js";
+import { ArtifactResourceScope } from "../artifacts/ArtifactResourceScope.js";
+
+const selectAgent = async (agent: string) => {
+  const root = await createTestTempDirectory("rea-mcp-alias-");
+  const resources = new ArtifactResourceScope();
+  try {
+    return await resolveMcpRegistrationClients(
+      {
+        agents: [agent],
+        global: true,
+        cwd: root,
+        home: root,
+        platform: process.platform,
+        environment: {},
+      },
+      resources,
+    );
+  } finally {
+    await resources.close();
+  }
+};
 
 it.each([
   ["cline-vscode", "cline"],
@@ -23,15 +44,22 @@ it.each([
   ["kiro", "kiro-cli"],
   ["mastra", "mastracode"],
   ["pi-agent", "pi"],
-])("retains helper alias %s as %s", (alias, expected) => {
-  expect(resolveMcpRegistrationClientId(alias.toUpperCase())).toBe(expected);
+])("resolves registration alias %s as %s", async (alias, expected) => {
+  const result = await selectAgent(alias.toUpperCase());
+  if (!result.ok) throw result.error;
+  expect(result.value.map(({ id }) => id)).toEqual([expected]);
 });
 
-it("does not treat prototype names or unknown strings as client identities", () => {
-  expect(resolveMcpRegistrationClientId("__proto__")).toBeUndefined();
-  expect(resolveMcpRegistrationClientId("constructor")).toBeUndefined();
-  expect(resolveMcpRegistrationClientId("unknown-client")).toBeUndefined();
-});
+it.each(["__proto__", "constructor", "unknown-client"])(
+  "rejects unknown registration identity %s",
+  async (agent) => {
+    const result = await selectAgent(agent);
+    expect(result).toMatchObject({
+      ok: false,
+      error: { _tag: "ConfigurationError" },
+    });
+  },
+);
 
 const clientSchema = z.object({
   id: z.string(),
@@ -256,133 +284,136 @@ const REVIEWED_CLIENT_IDS = [
   "amp",
 ];
 
-it("resolves selected-environment paths and scopes without external catalog readers", async () => {
-  const home = await createTestTempDirectory("rea-registration-catalog-");
-  const processes = createTestProcessRegistry();
-  try {
-    await mkdir(join(home, "config"));
-    const child = await processes.run(
-      process.execPath,
-      ["--import", "tsx", "--input-type=module", "--eval", CATALOG_PROGRAM],
-      {
-        timeoutMs: 50_000,
-        env: {
-          PATH: process.env.PATH,
-          SystemRoot: process.env.SystemRoot,
-          WINDIR: process.env.WINDIR,
-          TMPDIR: process.env.TMPDIR,
-          TMP: process.env.TMP,
-          TEMP: process.env.TEMP,
-          HOME: home,
-          USERPROFILE: home,
-          XDG_CONFIG_HOME: join(home, "config"),
-          XDG_DATA_HOME: join(home, "data"),
-          XDG_CACHE_HOME: join(home, "cache"),
-          XDG_STATE_HOME: join(home, "state"),
-          APPDATA: join(home, "appdata"),
-          LOCALAPPDATA: join(home, "localappdata"),
+it(
+  "resolves selected-environment paths and scopes without external catalog readers",
+  { tags: ["long-running"] },
+  async () => {
+    const home = await createTestTempDirectory("rea-registration-catalog-");
+    const processes = createTestProcessRegistry();
+    try {
+      await mkdir(join(home, "config"));
+      const child = await processes.run(
+        process.execPath,
+        ["--import", "tsx", "--input-type=module", "--eval", CATALOG_PROGRAM],
+        {
+          env: {
+            PATH: process.env.PATH,
+            SystemRoot: process.env.SystemRoot,
+            WINDIR: process.env.WINDIR,
+            TMPDIR: process.env.TMPDIR,
+            TMP: process.env.TMP,
+            TEMP: process.env.TEMP,
+            HOME: home,
+            USERPROFILE: home,
+            XDG_CONFIG_HOME: join(home, "config"),
+            XDG_DATA_HOME: join(home, "data"),
+            XDG_CACHE_HOME: join(home, "cache"),
+            XDG_STATE_HOME: join(home, "state"),
+            APPDATA: join(home, "appdata"),
+            LOCALAPPDATA: join(home, "localappdata"),
+          },
         },
-      },
-    );
-    expect(child.exitCode, child.stderr).toBe(0);
-    const result = catalogObservationsSchema.parse(JSON.parse(child.stdout));
-    assertKiloNativeLayers(result, home);
-    expect(result.empty.ok).toBe(true);
-    if (!result.empty.ok) throw new Error(result.empty.error.message);
-    expect(result.empty.value.map((client) => client.id)).toEqual(
-      REVIEWED_CLIENT_IDS,
-    );
-    expect(result.layeredKilo.ok).toBe(true);
-    if (!result.layeredKilo.ok)
-      throw new Error(result.layeredKilo.error.message);
-    const kiloLayers = [
-      "config.json",
-      "kilo.json",
-      "kilo.jsonc",
-      "opencode.json",
-      "opencode.jsonc",
-    ].map((name) => join(home, "layered", "kilo", name));
-    expect(result.layeredKilo.value[0]).toMatchObject({
-      configPath: kiloLayers.at(-1),
-      configPaths: kiloLayers,
-    });
-    expect(result.layeredOpenCode.ok).toBe(true);
-    if (!result.layeredOpenCode.ok)
-      throw new Error(result.layeredOpenCode.error.message);
-    expect(result.layeredOpenCode.value[0]?.configPaths).toEqual(
-      ["config.json", "opencode.json", "opencode.jsonc"].map((name) =>
-        join(home, "config", "opencode", name),
-      ),
-    );
-    expect(result.explicit.ok).toBe(true);
-    if (!result.explicit.ok) throw new Error(result.explicit.error.message);
-    expect(result.explicit.value.map((client) => client.id)).toEqual([
-      "codex",
-      "amp",
-      "gemini-cli",
-      "fx",
-      "windsurf",
-    ]);
-    expect(result.explicit.value[0]?.configPath).toBe(
-      join(home, "selected-codex", "config.toml"),
-    );
-    expect(
-      result.explicit.value.find((client) => client.id === "fx"),
-    ).toMatchObject({ format: "fx", privateProfile: true });
-    expect(result.detected.ok).toBe(true);
-    if (!result.detected.ok) throw new Error(result.detected.error.message);
-    expect(result.detected.value.map((client) => client.id)).toEqual([
-      "claude-code",
-      "cursor",
-      "amp",
-    ]);
-    expect(result.project.ok, JSON.stringify(result.project)).toBe(true);
-    if (!result.project.ok) throw new Error(result.project.error.message);
-    expect(result.project.value[0]?.format).toBe("copilot_cli_project");
-    expect(
-      result.project.value.map(({ configPath, scope }) => ({
-        configPath,
-        scope,
-      })),
-    ).toEqual([
-      {
-        configPath: join(home, "project", ".github", "mcp.json"),
-        scope: "project",
-      },
-      ...(process.platform === "win32"
-        ? []
-        : [
-            {
-              configPath: join(home, "project", ".kilocode", "kilo.jsonc"),
-              scope: "project",
-            },
-          ]),
-      {
-        configPath: join(home, "project", "opencode.jsonc"),
-        scope: "project",
-      },
-    ]);
-    for (const refused of [
-      result.onlyGlobal,
-      result.defaultProject,
-      result.projectAmp,
-      result.unknown,
-      ...(result.blockedCandidate === null ? [] : [result.blockedCandidate]),
-      result.unknownDetection,
-    ]) {
-      expect(refused.ok).toBe(false);
-      if (refused.ok) throw new Error("Expected preflight refusal.");
-      expect(refused.error.tag).toBe("ConfigurationError");
+      );
+      expect(child.exitCode, child.stderr).toBe(0);
+      const result = catalogObservationsSchema.parse(JSON.parse(child.stdout));
+      assertKiloNativeLayers(result, home);
+      expect(result.empty.ok).toBe(true);
+      if (!result.empty.ok) throw new Error(result.empty.error.message);
+      expect(result.empty.value.map((client) => client.id)).toEqual(
+        REVIEWED_CLIENT_IDS,
+      );
+      expect(result.layeredKilo.ok).toBe(true);
+      if (!result.layeredKilo.ok)
+        throw new Error(result.layeredKilo.error.message);
+      const kiloLayers = [
+        "config.json",
+        "kilo.json",
+        "kilo.jsonc",
+        "opencode.json",
+        "opencode.jsonc",
+      ].map((name) => join(home, "layered", "kilo", name));
+      expect(result.layeredKilo.value[0]).toMatchObject({
+        configPath: kiloLayers.at(-1),
+        configPaths: kiloLayers,
+      });
+      expect(result.layeredOpenCode.ok).toBe(true);
+      if (!result.layeredOpenCode.ok)
+        throw new Error(result.layeredOpenCode.error.message);
+      expect(result.layeredOpenCode.value[0]?.configPaths).toEqual(
+        ["config.json", "opencode.json", "opencode.jsonc"].map((name) =>
+          join(home, "config", "opencode", name),
+        ),
+      );
+      expect(result.explicit.ok).toBe(true);
+      if (!result.explicit.ok) throw new Error(result.explicit.error.message);
+      expect(result.explicit.value.map((client) => client.id)).toEqual([
+        "codex",
+        "amp",
+        "gemini-cli",
+        "fx",
+        "windsurf",
+      ]);
+      expect(result.explicit.value[0]?.configPath).toBe(
+        join(home, "selected-codex", "config.toml"),
+      );
+      expect(
+        result.explicit.value.find((client) => client.id === "fx"),
+      ).toMatchObject({ format: "fx", privateProfile: true });
+      expect(result.detected.ok).toBe(true);
+      if (!result.detected.ok) throw new Error(result.detected.error.message);
+      expect(result.detected.value.map((client) => client.id)).toEqual([
+        "claude-code",
+        "cursor",
+        "amp",
+      ]);
+      expect(result.project.ok, JSON.stringify(result.project)).toBe(true);
+      if (!result.project.ok) throw new Error(result.project.error.message);
+      expect(result.project.value[0]?.format).toBe("copilot_cli_project");
+      expect(
+        result.project.value.map(({ configPath, scope }) => ({
+          configPath,
+          scope,
+        })),
+      ).toEqual([
+        {
+          configPath: join(home, "project", ".github", "mcp.json"),
+          scope: "project",
+        },
+        ...(process.platform === "win32"
+          ? []
+          : [
+              {
+                configPath: join(home, "project", ".kilocode", "kilo.jsonc"),
+                scope: "project",
+              },
+            ]),
+        {
+          configPath: join(home, "project", "opencode.jsonc"),
+          scope: "project",
+        },
+      ]);
+      for (const refused of [
+        result.onlyGlobal,
+        result.defaultProject,
+        result.projectAmp,
+        result.unknown,
+        ...(result.blockedCandidate === null ? [] : [result.blockedCandidate]),
+        result.unknownDetection,
+      ]) {
+        expect(refused.ok).toBe(false);
+        if (refused.ok) throw new Error("Expected preflight refusal.");
+        expect(refused.error.tag).toBe("ConfigurationError");
+      }
+      expect(result.cancelled).toMatchObject({
+        ok: false,
+        error: { tag: "AnalysisCancelledError" },
+      });
+    } finally {
+      await processes.shutdown();
+      await rm(home, { recursive: true, force: true });
     }
-    expect(result.cancelled).toMatchObject({
-      ok: false,
-      error: { tag: "AnalysisCancelledError" },
-    });
-  } finally {
-    await processes.shutdown();
-    await rm(home, { recursive: true, force: true });
-  }
-}, 60_000);
+  },
+);
 
 const assertKiloNativeLayers = (
   result: z.infer<typeof catalogObservationsSchema>,

src/application/McpRegistrationCatalog.ts

@@ -356,7 +356,7 @@ const ALIASES = {
 const isClientId = (value: string): value is McpRegistrationClientId =>
   Object.hasOwn(MCP_REGISTRATION_CLIENTS, value);
 
-export const resolveMcpRegistrationClientId = (
+const resolveMcpRegistrationClientId = (
   input: string,
 ): McpRegistrationClientId | undefined => {
   const lower = input.toLowerCase();

src/application/MipsGnuAttributes.ts

@@ -1,3 +1,4 @@
+import { readElfAttributeUleb } from "../domain/elfAttributeUleb.js";
 import type { MipsGnuAttributes } from "../domain/binaryTargetTypes.js";
 import { err, ok, type Result } from "../domain/result.js";
 
@@ -22,21 +23,6 @@ export const parseMipsGnuAttributes = (
 
   const readU32 = (offset: number): number =>
     littleEndian ? bytes.readUInt32LE(offset) : bytes.readUInt32BE(offset);
-  const readUleb = (
-    offset: number,
-    end: number,
-  ): { readonly value: number; readonly next: number } | undefined => {
-    let value = 0;
-    for (let index = 0; index < 5 && offset + index < end; index += 1) {
-      const byte = bytes[offset + index]!;
-      value += (byte & 0x7f) * 2 ** (7 * index);
-      if ((byte & 0x80) === 0)
-        return value <= 0xffffffff
-          ? { value, next: offset + index + 1 }
-          : undefined;
-    }
-    return undefined;
-  };
 
   let fpAbi: number | null = null;
   let msa: number | null = null;
@@ -65,7 +51,7 @@ export const parseMipsGnuAttributes = (
     let cursor = vendorEnd + 1;
     while (cursor < end) {
       const scopeStart = cursor;
-      const scope = readUleb(cursor, end);
+      const scope = readElfAttributeUleb(bytes, cursor, end);
       if (scope === undefined)
         return err("malformed MIPS GNU attribute scope tag");
       if (scope.value !== TAG_FILE)
@@ -85,7 +71,7 @@ export const parseMipsGnuAttributes = (
       const scopeEnd = scopeStart + scopeLength;
       cursor = scope.next + 4;
       while (cursor < scopeEnd) {
-        const tag = readUleb(cursor, scopeEnd);
+        const tag = readElfAttributeUleb(bytes, cursor, scopeEnd);
         if (tag === undefined) return err("malformed MIPS GNU attribute tag");
         if (
           tag.value !== TAG_GNU_MIPS_ABI_FP &&
@@ -94,7 +80,7 @@ export const parseMipsGnuAttributes = (
           return err(
             `MIPS ELF declares GNU attribute tag ${tag.value}, which REA does not inspect`,
           );
-        const value = readUleb(tag.next, scopeEnd);
+        const value = readElfAttributeUleb(bytes, tag.next, scopeEnd);
         if (value === undefined)
           return err("malformed MIPS GNU attribute value");
         if (tag.value === TAG_GNU_MIPS_ABI_FP) {

src/application/MsixPackageStorage.ts

@@ -54,7 +54,7 @@ const pathWithin = (path: string, root: string): boolean => {
  * processes see those files at virtual AppData locations; other processes do
  * not, and Windows removes them with the package.
  */
-export const packagePrivateStorageFamily = (
+const packagePrivateStorageFamily = (
   canonicalPath: string,
   packagesDirectory: string,
 ): string | undefined => {

src/application/OperationAlternatives.ts

@@ -0,0 +1,84 @@
+import type { AnalysisOperation } from "./AnalysisProvider.js";
+
+/** One operation that observes part of what a missing operation answers. */
+interface OperationAlternative {
+  readonly operation: AnalysisOperation;
+  /** What it returns, as a noun phrase. */
+  readonly observes: string;
+}
+
+/**
+ * Follow-ups for one question shape that an undeclared operation would have
+ * answered. They are conditional rather than substitutes: `limitation`
+ * states what none of them establishes.
+ */
+interface OperationAlternativeRelation {
+  readonly question: string;
+  readonly alternatives: readonly OperationAlternative[];
+  readonly limitation: string;
+}
+
+const INSTRUCTION_OBSERVATIONS: OperationAlternativeRelation = {
+  question: "instruction-level questions",
+  alternatives: [
+    {
+      operation: "inspect_native_instruction",
+      observes:
+        "the decoded instruction or data at one exact address, with its length and static flow",
+    },
+    {
+      operation: "read_function_instructions",
+      observes: "one analyzed procedure's instruction listing in address order",
+    },
+    {
+      operation: "procedure_assembly",
+      observes: "one analyzed procedure's assembly text",
+    },
+  ],
+  limitation:
+    "None navigates arbitrary analyzed objects such as data or addresses outside a procedure, and neither listing order nor address plus length proves executed control flow.",
+};
+
+/** Operations absent from this relation keep their consumer's generic advice. */
+const OPERATION_ALTERNATIVES: ReadonlyMap<
+  string,
+  OperationAlternativeRelation
+> = new Map<AnalysisOperation, OperationAlternativeRelation>([
+  ["next_address", INSTRUCTION_OBSERVATIONS],
+  ["prev_address", INSTRUCTION_OBSERVATIONS],
+]);
+
+/** Declared alternatives the caller can run now, in declaration order. */
+export interface OfferedOperationAlternatives {
+  readonly operations: readonly AnalysisOperation[];
+  /** Each offer's scope, what none establishes, and the provider choice. */
+  readonly remediation: string;
+}
+
+/**
+ * Select the declared alternatives for an operation the current provider does
+ * not declare. `isOffered` holds only for operations that provider makes
+ * available and the caller advertises. Undefined means none qualifies.
+ */
+export const offeredOperationAlternatives = (
+  operation: string,
+  isOffered: (alternative: AnalysisOperation) => boolean,
+): OfferedOperationAlternatives | undefined => {
+  const relation = OPERATION_ALTERNATIVES.get(operation);
+  if (relation === undefined) return undefined;
+  const offered = relation.alternatives.filter((alternative) =>
+    isOffered(alternative.operation),
+  );
+  if (offered.length === 0) return undefined;
+  const described = offered.map(
+    ({ operation: alternative, observes }) => `${alternative} (${observes})`,
+  );
+  const listed =
+    described.length === 1
+      ? described[0]
+      : `${described.slice(0, -1).join(", ")} and ${described.at(-1)}`;
+  return {
+    operations: offered.map((alternative) => alternative.operation),
+    remediation: `For ${relation.question}, the current provider offers ${listed}. ${relation.limitation} Running ${operation} itself requires a provider that declares it.`,
+  };
+};

src/application/ReferenceSourceImportEntries.ts

@@ -34,7 +34,7 @@ export interface ParsedReferenceSourceEntries {
 }
 
 /** Preserve the observed reader failure and add entry-specific recovery guidance. */
-export const projectReferenceSourceEntryFailure = (
+const projectReferenceSourceEntryFailure = (
   entry: Extract<ReferenceSourceEntry, { status: "failed" }>,
 ): string => `${entry.message} ${entryRecoveryGuidance(entry)}`;
 

src/application/ReferenceSourceImportTypes.test.ts

@@ -1,7 +1,6 @@
 import { describe, expect, it } from "vitest";
 
 import {
-  projectReferenceSourceImportError,
   referenceSourceImportAnalysisError,
   type ReferenceSourceImportError,
 } from "./ReferenceSourceImportTypes.js";
@@ -20,11 +19,13 @@ describe("reference-source import error projection", () => {
       parse: "execution_failure",
     } as const;
     for (const [code, category] of Object.entries(expectedCategories)) {
-      const projected = projectReferenceSourceImportError({
-        tag: "reference-source-import",
-        code: code as ReferenceSourceImportError["code"],
-        message: "EIO while reading /selected/source/src/main.ts",
-      });
+      const projected = projectAnalysisError(
+        referenceSourceImportAnalysisError({
+          tag: "reference-source-import",
+          code: code as ReferenceSourceImportError["code"],
+          message: "EIO while reading /selected/source/src/main.ts",
+        }),
+      );
       expect(analysisErrorProjectionSchema.safeParse(projected).success).toBe(
         true,
       );

src/application/ReferenceSourceImportTypes.ts

@@ -5,10 +5,6 @@ import {
   AnalysisOutputError,
 } from "../domain/analysisErrorCore.js";
 import { analysisErrorWithCancellation } from "../domain/analysisErrorCancellation.js";
-import {
-  projectAnalysisError,
-  type AnalysisErrorProjection,
-} from "../domain/analysisErrorProjection.js";
 import type {
   AnalysisError,
   AnalysisErrorOptions,
@@ -125,12 +121,6 @@ export const referenceSourceImportAnalysisError = (
       );
 };
 
-/** Canonical CLI/MCP projection, retaining source facts under diagnostic details. */
-export const projectReferenceSourceImportError = (
-  error: ReferenceSourceImportError,
-): AnalysisErrorProjection =>
-  projectAnalysisError(referenceSourceImportAnalysisError(error));
-
 /** Import a caller-selected local source tree as historical reference. */
 export interface ReferenceSourceImportOptions {
   readonly root: string;