mirror of
https://github.com/Twigpine/zero.git
synced 2026-11-11 13:11:30 -10:00
fix/undo-command
70
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 Vasanth T
|
378d538e24 |
perf(agent): execution profiles with one-shot posture escalation (PR10b+PR10c) (#740)
* perf(agent): execution profiles, --exec-profile, bench passthrough, TUI /profile (PR10b+10c) * fix(execprofile): harden profiles against review findings - fast escalates on critical-risk mutations, not high: the sandbox marks every shell command high before command analysis, so a high trigger spent the one-shot on the first go test of any task - escalation can now restore the provider-default effort a profile displaced (PostureEscalation.RestoreDefaultEffort; a plain empty target means leave untouched) - TUI /turns while a profile is active pins the budget: disarms the escalation turn target (mirroring exec's explicit --max-turns rule) and profile reverts use touched bits instead of value equality, so explicit choices survive even when they coincide with profile values - spec-draft runs get the same escalation policy as the main run instead of the displaced budget with no safety net - bench validates and canonicalizes --exec-profile before spawning: an unknown name previously produced an exit-0 report full of millisecond usage-error walls - turn summary prints exec-profile alongside mode; zero exec rejects an inline-empty --exec-profile=; balanced trace label documented and pinned as deliberate attribution * fix(execprofile): explicit effort choices disarm the escalation effort restore An explicit effort set AFTER profile selection could be cleared by a mid-run escalation (the effort analog of the /turns pin, introduced by RestoreDefaultEffort): TUI /effort and Ctrl+T now disarm the restore on a policy clone, and the spec-draft path only arms it when no explicit --spec-reasoning-effort governs the draft. * fix(execprofile): address review findings on the profile surfaces - --use-spec projects away a profile's self-correct knob instead of silently arming an option the draft runner ignores (the explicit --self-correct --use-spec combination is already a parse-time error) - RunTurnBench canonicalizes and validates ExecProfile at the library boundary, so a direct caller's casing/whitespace cannot fork report labels and an unknown name fails before anything runs - reverting a profile whose displaced budget was zero clears ZERO_MAX_TURNS (SetMaxTurnsEnv ignores zero, so sub-agents would have kept the removed profile's budget) - the TUI effort-disarm test fails instead of skipping when the catalog stops filling low effort * fix(tui): reconcile an active profile's effort fill on model switch The fill is per-model, not a session preference: fast selected on a model without low fills nothing, and switching to a model that supports it should behave like selecting the profile there; the reverse switch must not keep sending a level the destination does not support. Both switch paths (the /model command and the cross-provider picker) now re-derive the fill, leaving explicitly touched efforts alone and keeping the escalation's effort restore in step. Also: the revert-env test asserts key removal via LookupEnv, and applyProfileTurnBudget's pinned-budget parameter is named and documented as covering mode presets, not just the flag. * fix(tui): void profile effort bookkeeping when the model-switch drop erases it The /model path's pre-existing unsupported-effort drop ran before the profile reconciliation and left contradictory state behind: a dropped touched override kept blocking the profile from governing, and a dropped profile fill left a stale applied value with the effort restore still armed. The drop itself is load-bearing (the TUI run path forwards the effort to the provider unfiltered), so instead of preserving a known-unsupported level, the voided choice now resets the profile's effort bookkeeping and the reconcile re-derives from scratch. The whole sequence lives in one ring-parameterized helper driven by the switch target's authoritative ring, unit-tested against synthetic rings for both the void-then-refill arc and the supported-override survival. * fix(tui): only drop an explicit effort when the target ring is authoritative A nil ring meant both a catalog model with no effort controls and a live-discovered custom model the catalog cannot vouch for. The drop now requires an authoritative ring (a resolved catalog entry, whose ring may legitimately be empty); an unknown ring preserves an explicit preference, matching the cross-provider picker's behavior, while the profile's own fill still applies only where support is known. |
||
|
Vasanth T
|
af875df587 |
perf(agent): posture-escalation signals and controller (PR10a) (#736)
* trace(perf): add posture escalation counter * feat(agent): typed posture-escalation policy and tool-result risk field * feat(agent): posture controller wiring and executed-risk stamping * test(agent): cover posture controller, escalation act path, and risk stamping * test(agent): pin zero risk on not-executed results * fix(agent): wire uncertain-path escalation and validate risk thresholds |
||
|
Vasanth T
|
30e2c3f7ff |
perf(providers): provider capabilities and default turn-session adapter (PR7) (#720)
* perf(providers): add provider capabilities and default turn-session adapter (PR7) * fix(providers): project effective reasoning efforts into capabilities * feat(agent): carry optimized turn sessions across mid-run model switches * test(agent): cover swap-time session open failure |
||
|
Anandan
|
18cce358a2 |
Add deterministic completion policy (#719)
Extract completion decisions from the agent loop into a feature-gated typed policy with complete, incomplete, and uncertain outcomes. Preserve bounded plan-stall nudges and allow at most one semantic acceptance check for self-correcting runs. Tested: make build Tested: make test Tested: go fmt ./... Tested: go vet ./... Tested: make lint Tested: govulncheck ./... Note: repository-wide pinned golangci-lint reports 36 pre-existing unrelated findings; internal/agent/... reports 0 issues. |
||
|
KRATOSandKRATOS
|
e5670c427f |
perf(output): add token-aware semantic output budgeting (PR11) (#717)
* perf(output): add token-aware budget contract * perf(output): add semantic retention policies * perf(output): budget results after redaction * perf(output): propagate tool truncation metadata * perf(output): trace semantic budget decisions * perf(output): classify core tool output * docs(output): describe semantic output budgeting * test(output): make spill fallback portable on windows * fix(output): address semantic budgeting review findings * test(output): make hook budgeting tests portable * fix(output): retain oversized single-line file content * fix(output): apply semantic budgets to captured output * fix(output): preserve hook budgets and search paths * fix(output): preserve alias and truncation metadata * test(trace): validate output budget metadata --------- Co-authored-by: KRATOS <[email protected]> |
||
|
KRATOSandKRATOS
|
31d45d5f14 |
perf(agent): concurrent read-only tool batches via capability gate (#715)
Upgrade the parallel tool planner from SideEffectRead probes to the PR5 CapabilitiesOf contract: EffectReadOnly + ThreadSafe + auto-allowed, with resource-key conflict boundaries so same-path reads stay sequential. Mark audited pure reads ThreadSafe (read_file, read_minified_file, list_directory, glob, grep, skill, web_fetch). Glob/grep use scopedScan directory keys (nil for workspace-wide "." so scans do not false-conflict). Tests cover the capability gate, key conflicts, extendParallelRun windows, and catalog ThreadSafe audit for concurrent-safe reads. Co-authored-by: KRATOS <[email protected]> |
||
|
PierrunoYTandClaude Sonnet 5
|
824ecdbcf9 |
fix(hooks): run sessionEnd hooks after Esc/Ctrl+C interrupts (#606)
* fix(hooks): run sessionEnd hooks after Esc/Ctrl+C interrupts On interrupt, Run returns with ctx already canceled. The deferred dispatchSessionEnd call passed that same ctx into Hooks.Dispatch, whose context.WithTimeout derives an already-canceled child context, so the hook process never actually launches (recorded as timed-out/error in the audit log). Since most sessions end via interrupt, sessionEnd hooks were effectively unreliable. Detach the sessionEnd dispatch from the run context's cancellation with context.WithoutCancel so the dispatcher's own timeout still applies without being short-circuited by the run's cancellation. Co-Authored-By: Claude Sonnet 5 <[email protected]> * test(agent): fall back to GOROOT for hook regression * test(agent): justify GOROOT fallback --------- Co-authored-by: Claude Sonnet 5 <[email protected]> |
||
|
Anandanandanandh8x
|
1c5c6e78a8 |
feat(perf): emit prompt-prefix hash fingerprint per turn (#704)
Compute a seven-field SHA-256 fingerprint of the cacheable prompt prefix (base instructions, confirmation policy, project context, skills, tools, tool schemas, complete) and emit one prefix_hash event per turn through the trace recorder. The complete hash is the SHA-256 of the canonical join of the other six, so any sub-component drift is observable both individually and in aggregate. The decomposition exists so a downstream observer can detect which sub-component of the prefix broke cache stability across turns: a stable complete_prefix hash means the cacheable prefix is byte- identical, and the sub-hash that changed names the culprit. No behavior change for untraced runs. For OpenAI, the existing prompt_cache_key (session ID) is unchanged; the fingerprint is purely additive observability. The Anthropic cache_control breakpoint work already in main is not touched. Co-authored-by: anandh8x <[email protected]> |
||
|
Vasanth T
|
abe24bcb10 |
Add per-turn tracing and a turn-benchmark harness (Phase 0 baseline) (#700)
* Add per-turn tracing and a turn-benchmark harness (Phase 0 baseline)
Introduce an opt-in `internal/trace` package that attributes a run's wall
time to named spans — prompt build, provider connect/queue, generation,
tool queue/execution, permission wait, verification, compaction,
persistence — plus counters for model requests, tool calls, retries,
reconnects, compactions, completion nudges, acceptance checks, model
switches, and input/cached/output tokens. The recorder is nil-safe (a nil
*Recorder is a no-op), so every stamp site can call it unguarded; an
untraced run is byte-identical to today.
The recorder threads into agent.Run via a new `Options.Trace` field and a
single `trace.WithContext` assignment, so the shared providerio seam
(connect around client.Do, queue around OAuth resolve, first token at the
first non-keepalive SSE payload) reaches it through `trace.FromContext`
without any provider-interface change. All instrumentation is opt-in: when
`Options.Trace` is nil the hot path is untouched.
Headless `zero exec` gains `--trace <path>` (and `ZERO_TRACE=<path>`),
writing an agenteval-contract-compatible NDJSON trace (`-` for stderr).
The trace is pure observation; enabling it does not change behavior.
A new per-turn benchmark harness extends `internal/perfbench` (reusing the
existing TaskSet/NumericStats/subprocess runner shapes): `RunTurnBench`
aggregates per-span median/P95 across iterations, ranks the top three
controllable latency sources by share of attributed time, rolls up per
task class, and totals tokens/counts. A checked-in 42-task manifest across
seven classes (nav, edit, fix, refactor, longproc, longctx, parallel) and
local offline fixtures back it, and `make baseline` runs it against the
built `zero`, writing `internal/perfbench/reports/baseline.json`. The
report is machine-specific and regenerated, not hand-edited.
The manifest is JSON rather than YAML to avoid adding a dependency;
per-task verification targets the specific test (`go test -run <name>`)
so fix tasks are independent. Each task is a fresh `zero exec` process, so
iterations are cold-start samples; a warm in-process path is left for
follow-up.
Verified: `go build ./...`, `go vet ./...`, `gofmt -l` clean on touched
files; `go test -race` green for internal/trace, internal/agent,
internal/perfbench, and cmd/zero-perf-bench.
* test(trace): make TestAttributionRatio clock-independent
The test recorded two synthetic 10ms spans with no real delay and asserted
the attribution ratio stayed <= 1.0. On a precise clock (Linux/macOS CI) the
wall is microseconds while attributed is 20ms, so attributed/wall blows past
1.0 and the test failed — even though the ratio legitimately exceeds 1.0
when spans overlap (parallel tool execution). On Windows the coarse timer
masked it, so the failure only surfaced in CI.
Assert the contract instead: attributed is the deterministic sum of span
durations, and the ratio equals attributed/wall (0 when wall is zero). Add a
zero-wall case for the divide-by-zero guard.
* Address review feedback on the per-turn tracer + benchmark harness
Benchmark runner (internal/perfbench):
- A nonzero agent exit code no longer counts as Passed. The runner set
VerifyErr but then fell through to outcome.Passed = true, so a crashed
run could be recorded as a pass. Return before the verification block
when VerifyErr is already set.
- A task now passes only when every iteration passes. The previous
"any iteration passes" rule let a flaky fail hide behind one good run.
- Each invocation copies its workspace fixture to a fresh temp dir and
runs the agent + verifier there, so mutating tasks (edit/fix/refactor)
can't dirty the checked-in fixtures or bleed into the next iteration.
- Trace load failures (missing/malformed trace file) are surfaced as
result warnings via a new TraceIssue field instead of being swallowed.
Tracer (internal/trace):
- Finish freezes the recorder: Counter, addSpan, and the first-event
stamps are dropped after Finish returns, so a snapshot captures exactly
one turn and late stamps can't mutate it.
- WriteText propagates the first write error instead of silently
swallowing sink failures.
Provider seam (internal/providers/providerio):
- FirstTokenAt is stamped only after a payload is accepted as real model
output, not before handle() decides to drop it as an error payload.
Agent loop (internal/agent):
- The compaction counter increments only when the history actually
shrinks, in both the proactive and reactive paths — paid no-ops no
longer inflate the compaction total.
- Document on Options.Trace that a fresh Recorder is required per Run.
CLI (internal/cli/exec):
- Finish the trace at the agent.Run boundary and defer only the
serialization, so the snapshot covers exactly the turn.
- Reject --trace / ZERO_TRACE for spec-draft runs with a clear error
instead of silently accepting and writing nothing.
Manifest + fixtures:
- Strengthen the edit verifiers: rename and value-bump tasks now assert
the old symbol is gone, and the getter task greps the method signature.
- Add the missing testdata/edit fixture (10 edit tasks referenced a
directory that was never committed, so the whole edit class errored
out at run time).
- The manifest test now stats each fixture so a missing dir fails at
load time instead of at every run.
- longproc benchmark keeps Process live across the loop; README code
fence gets a language tag and documents the per-invocation fixture copy.
- Add tests for the tracing wrapper (nil vs wired recorder, OnUsage
forwarding, token aggregation) and the frozen-recorder contract.
* Address review feedback: exclusive-time attribution + wired spans + fix-suite compile
Rewrite the trace attribution model so concurrent/nested spans no longer
double-count. Spans record wall intervals; the recorder derives a parent for
each by interval containment and computes exclusive time (duration minus the
union of its children). AttributedDuration sums only top-level spans; Coverage
is the union of intervals over wall, capped at 1, the honest ">=95%" metric.
Top-latency ranking uses exclusive time so shares sum to ~1 instead of being
inflated by overlapping provider_connect/generation and nested
permission_wait/tool_execution.
Wire the spans that were named but never stamped: verification around the
self-correct verify pass, and StampFirstVisibleEvent on the first forwarded
text event. Drop the tool_queue/process_wait/persistence constants that had no
stamps, and rename the misnamed near-noise prompt_build span to tool_partition.
Trim RequiredEventKeys to events a healthy short run actually emits; move
conditional phases (tools, permission, compaction, verification) to
OptionalEventKeys so a short trace does not false-fail.
Make ReadNDJSON fail loudly on a corrupt file: require a type:trace header,
error on span/counter lines that appear before it, and error when a header
yields no spans or counters.
Fix the blocking fix-suite compile error: Label's %d format verb on a string
tripped go test's vet before any fix task could verify. Make it a runtime
wrong-prefix bug ("admin-" vs "user-") that still type-checks, so each fix
task is independent again.
Manifest: correct refactor-04's stats.go to main.go, reword fix-07 to the
wrong-prefix bug, add a latency-only description (read-only tasks carry no
verification oracle by design), and frame the report README on exclusive time
and coverage.
Tests updated for the occurrence/exclusive/coverage model; trace, perfbench,
and agent green under -race.
* Fix exclusive-time round-trip, identical-interval nesting, and empty-trace parse
Three correctness bugs in the attribution model found by an adversarial review
of the prior commit, all undermining the no-double-count guarantee:
1. exclusive_ms:0 was overwritten with the inclusive Duration on re-parse. A
parent whose children tile its interval legitimately has exclusive time 0;
WriteNDJSON emits exclusive_ms:0, but ReadNDJSON's "<= 0 means missing"
fallback turned that back into the full duration, so every re-parsed trace
re-introduced the double-counting the model exists to prevent (the benchmark
harness, which ranks re-parsed traces, inherits it). Gate the fallback on
key ABSENCE, so a written 0 is preserved.
2. deriveNesting formed a 2-cycle for two spans with identical [start, end]
intervals: the symmetric containment check made each parent the other,
dropping both from top-level and zeroing AttributedDuration. Break the tie
deterministically — strict containment is an unambiguous parent; for
identical intervals, only the lower-indexed span may parent the other, so
one stays top-level.
3. ReadNDJSON returned a valid empty TurnTrace for empty/blank-only input, so a
run that crashed before emitting a trace (or where --trace was not honored)
masqueraded as a clean zero-attribution sample. Empty input now errors, so
the harness records a TraceIssue.
Also: decode counters via json.Number so int64 values above 2^53 round-trip
exactly instead of losing precision through float64; and drop the harness's
own "exclusive <= 0 -> Duration" ranking fallback, which was the same
double-count on the re-parsed path. Added regression tests covering the
zero-exclusive round-trip, the identical-interval nesting, empty-input
rejection, and counter precision.
* perf(turn-bench): split pass/fail into oracle tiers so the pass rate can't be misread as correctness
The turn benchmark counted any exit-0 task as passed, including the 24
read-only tasks (nav/longproc/longctx/parallel) that carry no verification
oracle — so the JSON tasksPassed / "N/M passed" headline could read as a
correctness verdict it never was. Split pass/fail into three oracle tiers:
- correctness (edit grep + fix go test): the only pass rate that can move
with model quality (tasksVerified / tasksPassed / correctnessPassRate)
- build-only (refactor go build): a non-positive oracle, reported as
buildPassRate and excluded from correctnessPassRate; the manifest declares
buildOnlyClasses: ["refactor"]
- latency-only (no oracle): counted in latencyOnlyTasks, never in any pass
rate
NewTurnExecRunner no longer sets Passed=true for tasks with no
verificationCommand, and the summary headline names each tier explicitly so
an exit-0 read-only run cannot inflate a correctness number. Bumps
TurnSchemaVersion to 2. Class breakdown and tier contract documented in
internal/perfbench/MANIFEST.md and reports/README.md.
Also install the trace first-token/visible-event forwarding handlers in
agent.Run whenever options.Trace is set, not only when a UI OnText /
OnReasoning callback is set, so a headless traced run (zero exec --trace)
still captures FirstTokenAt. forwardedVisibleText stays tied to the user
callback so stall-retry semantics are unchanged. Adds a regression
assertion to the existing traced-run test.
* Address CodeRabbit review on
|
||
|
PierrunoYT
|
6ec77d62f5 | Dispatch session lifecycle hooks (#581) | ||
|
Pengkuh Dwi Septiandi
|
3f0503bc23 |
fix(agent): keep tools exposed for max-turn finalization (#533)
* fix(agent): keep tools exposed for max-turn finalization * chore(agent): rename max-turn tool definitions parameter |
||
|
Kevin Codex
|
95ccd5bc32 |
perf: universal tool-output ceiling with spill + async post-edit diagnostics (#518)
* feat(tools): universal token-based output ceiling with spill-to-disk
Cap every tool result at the registry boundary (16k estimated tokens,
ZERO_TOOL_OUTPUT_CEILING_TOKENS-tunable, 0 disables) unless the tool
manages its own deliberate budget. Head+tail truncation, full output
spilled to the existing redacted spill dir and re-readable via
grep/read_file, so nothing is lost — only deferred.
This closes the three unbounded paths a single call could flood the
context window through (and then re-bill every following turn):
- web_fetch: up to 256KiB default / 2MiB max of non-HTML body
- skill: full SKILL.md returned uncapped
- MCP-served tools: no budget at all
Self-budgeting tools opt out via an unexported marker interface (an MCP
server can never exempt itself): bash, exec_command (model-raisable),
read_file, read_minified_file, grep, glob, list_directory. Browser
snapshots deliberately stay under the net.
bash itself is tightened to match the unified token model: emit budget
96KiB -> 32KiB per stream (~24k -> ~8k tokens worst case per stream),
while capture retention stays at 96KiB head+tail so the new spill file
covers 6x more of the output than the transcript shows.
* perf(agent): collect post-edit LSP diagnostics off the tool-call path
Inline post-edit diagnostics blocked every edit_file/write_file result on
the language server: a >=300ms quiet-period debounce that reset on each
publish, seconds on the first edit per language, capped at 10s — the
largest recurring latency tax in an edit-heavy session.
Diagnostics are now collected in the background: a file is enqueued the
moment a mutating tool finishes, a single worker checks it while the rest
of the batch executes, and the loop drains completed results just before
building the NEXT provider request, appending errors as a user nudge.
The model always issues another request after an edit turn (to read its
tool results), so it still sees introduced errors at the same decision
point — no tool call ever stalls on the language server.
A file re-edited before its check runs is checked once in its final
state (the checker re-reads the file); re-edited after, it is re-queued
and the newer result wins. A drain that finds the worker still busy
gives up after 3s and delivers on a later turn — results stay accurate
because any further edit re-enqueues the file.
The tools-side inline mechanism (tools.RunOptions.Diagnostics) is kept
for direct API callers; the loop just no longer wires it.
* fix: address review — max-turns diagnostics drain, capture-gap marker, test hardening
- Drain background diagnostics before the max-turns final-answer request
so an error introduced by the LAST turn's edit is reported in the
summary instead of silently dropped (+ regression test).
- spillBashStreams marks the head/tail junction with a capture-gap
marker when boundedBuffer dropped the middle of a stream, so a spilled
log never reads as contiguous when it is not; junction position is
asserted in the test.
- Tests override TMP/TEMP alongside TMPDIR (os.TempDir reads TMP/TEMP on
Windows) via a shared setTestTempDir helper.
- TestBudgetBashCaptureReportsTrueTotal now asserts spill_path meta and
spill file content; exemption-list test gains read_minified_file.
* fix(test): use a real workspace root in async diagnostics tests
filepath.IsAbs("/ws") is false on Windows (no drive letter), so the
collector test's workspace-absolute assertion failed on the windows
smoke job. Root the fake workspace at t.TempDir() — absolute on every
platform — and assert against it.
* fix: pattern-scrub spill files and gate finalization on pending diagnostics
Review follow-ups (Vasanthdev2004):
- Spill files now pass the pattern-based secret scanner (AWS keys,
GitHub/Slack/OpenAI tokens, PEM blocks, JWTs) in addition to the
configured-key scrub. A bash spill runs before formatBashOutput's
scan, so without this a spilled file held in cleartext exactly the
credentials the transcript redacts. Applied centrally in
spillTruncatedOutput, covering bash, exec_command, and ceiling spills.
- Finalization diagnostics gate: a no-tool-call final answer now drains
pending post-edit diagnostics with the full inline-era budget (10s,
vs the 3s per-turn wait) before the run is accepted; introduced
errors append the nudge and give the model one more turn instead of
being silently dropped when no later turn exists. The max-turns
summary path uses the same finalization budget. Free for runs that
never edited — an idle collector returns immediately.
|
||
|
Kevin Codex
|
87e7e69afd |
feat(openai): forward prompt_cache_key for server-side prefix cache routing (#515)
* feat(openai): forward prompt_cache_key for server-side prefix cache routing Send the session ID as OpenAI's prompt_cache_key on every completion request so consecutive turns route to a backend replica that already holds the conversation prefix in its prompt cache, instead of re-billing the full ~11k-token system+tools prefix each turn on OpenAI-compatible endpoints (which get no cache_control breakpoints, unlike Anthropic). Keyless requests serialize byte-identically to before, and ZERO_DISABLE_PROMPT_CACHE_KEY=1 suppresses the field for strict endpoints that reject unknown parameters. The Codex (Responses API) flavor is deliberately untouched: the ChatGPT backend is a separate, stricter surface. * fix(openai): parse prompt-cache-key kill switch as a boolean flag ZERO_DISABLE_PROMPT_CACHE_KEY=0 or =false previously disabled forwarding because any non-empty value tripped the check. Parse it the same way ZERO_FORMAT_ON_WRITE parses boolean flags, so explicitly-falsy values are a no-op and only truthy values flip the toggle. |
||
|
Kevin Codex
|
3c81fea228 |
feat: agent quality, caching, retry, and tooling upgrades (#506)
* feat: agent quality, caching, retry, and tooling upgrades Ten self-contained improvements to the core agent engine and tools: - edit_file: fuzzy fallback matching (7-strategy cascade with uniqueness protection and a disproportion guard) when old_string does not match byte-for-byte; exact match remains the fast path - edit_file/write_file: inline error-severity LSP diagnostics in tool output so the model sees a break it introduced in the same turn - lsp: expand server registry from 4 to 28 languages; doctor grades only tier-1 servers and lists the long tail informationally - agent: consecutive read-only tool calls execute concurrently (bounded at 8), consumed in original order, never spanning a mutating call - anthropic: cache_control breakpoints on the last two conversation messages so the transcript prefix stays a prompt-cache hit - providerio: retry schedule raised to 6 attempts with exponential 2s-30s backoff when the server sends no Retry-After - tools: truncated exec/bash output spills to a redacted temp file referenced in the truncation notice; bash output (previously unbounded) now shares exec_command's token budget - web_fetch: dependency-free HTML-to-markdown conversion (format: auto|raw|markdown) and larger limits (256KiB default, 2MiB max, 30s) - modelregistry: live models.dev overlay refreshes context limits and base pricing from a TTL-cached fetch; opt-in at the CLI entrypoint - tools: opt-in format-on-write (ZERO_FORMAT_ON_WRITE=1) runs standard formatters across 27 extensions, ordered before the file-tracker re-baseline so conflict detection is preserved * fix: address PR review feedback on fuzzy edits, diagnostics, spill - edit_file: adapt the replacement to the fuzzy-resolved span — when the span is a uniformly deeper-indented copy of old_string (line-trimmed / indentation-flexible matches), the same indentation delta is applied to new_string, and a span ending in a bare CR keeps its CR so CRLF files never gain a mixed line ending; block-anchor matches with drifted interiors are deliberately left untouched - diagnostics: render workspace-relative paths in inline post-edit diagnostics (base name outside the workspace) so absolute paths never leak the local username or directory layout into the model prompt or session transcript - spill: harden the spill directory for shared temp dirs — per-uid name, reject symlinked or non-directory paths (MkdirAll follows symlinks), and on Unix require current-user ownership - bash: the branch's output cap and truncation-flag threading are superseded by main's capture-time bounding (boundedBuffer + budgetBashCapture), which already sets Result.Truncated; this rebase adopts that design and keeps spill-to-disk on exec_command, where the full output is still buffered and a spill is genuinely complete * fix: address review round 2 — permission-event race, fuzzy ambiguity, spill reachability - agent: serialize OnPermission behind the same mutex as OnPermissionRequest in the parallel read batch. buildPermissionEvent fires for batched reads with a non-trivial sandbox decision (e.g. a grant-matched extra root), and front-end handlers append to shared session-recording state without their own locking — two batched reads emitting concurrently was a data race the pre-batch serial loop never allowed. - edit_file: reject two or more DISTINCT candidate spans from one fuzzy strategy as ambiguous instead of silently editing the first. Duplicate blocks at different indentation each resolve to a unique literal, so the previous occurrence-uniqueness check passed and wrote the wrong span. A single candidate occurring at multiple positions still cascades to stricter strategies as before. - tools: make spill files actually reachable by the recovery path the truncation notice advertises — resolveScopedReadPath now recognizes files inside the per-uid spill dir (containment verified after symlink resolution, so a planted link cannot smuggle an out-of-scope file through the gate); read_file/grep can follow the notice instead of re-running the command. * fix: surface competing block-anchor spans; make write-barrier test observe cross-probe order - edit_file: blockAnchorReplacer now returns every span clearing the similarity threshold (best first) instead of silently picking the single most-similar block. Hiding competitors from the cascade's distinct-candidate check could edit the wrong block when two blocks share anchors and both interiors are plausible; one clear winner still resolves, two plausible blocks now report ambiguity (covered by a two-block test that asserts the file is untouched). - agent tests: the parallel write-barrier test now records both probes into one mutex-guarded shared log and asserts the cross-probe ordering it always claimed to (write starts after the first read batch ends; second read batch starts after the write ends). The previous version only compared indices within the read probe's own log, which a regression letting the write overlap a batch could still pass; the stray no-op mutex assignment is gone. |
||
|
Anandan
|
afb5010a46 |
polish v2: smoothness & delight (#484)
* feat(tui): actionable hints on provider/model errors Classify a failed turn's provider error into auth / rate-limit / connectivity / model-not-found / context-overflow and render a one-line next step below the red error row, instead of dumping an identical raw blob for every failure mode. The classifier lives in a new internal/errhint package so both the TUI error row and the CLI exec path can share it (TUIHint references slash commands, CLIHint references zero subcommands). Classification is a conservative string heuristic keyed off providerio.ClassifiedError's prefixes plus lower-level DNS/TLS/timeout/context-length signatures, since the numeric HTTP status is gone by the time the error reaches a UI surface. * feat(tools): budget bash output with head+tail truncation bash was the one tool that returned command output with no byte cap — 'cat large.log' or a verbose test run could dump megabytes into context and force compaction, while every other read/search tool already applies a budget. Cap stdout and stderr at 96KB each, keeping the head and tail of an oversized stream (build/test failures usually land at the tail) and dropping the middle behind a marker that points at redirect-to-file + read_file. Record raw_bytes/emitted_bytes/estimated_tokens/truncated in Meta like the other tools. Sandbox-denial and shell-issue detection still run on the full raw output before budgeting. * fix(gemini): retry 401 with refreshed OAuth token The Gemini provider used providerio.SendWithRetry (429/503/529 only), so an expired OAuth token surfaced as a terminal 'API key not valid' auth error while OpenAI and Anthropic silently recovered via SendWithAuthRetry's force-refresh-and-replay on 401. Plumb OAuthResolver through gemini.Options/Provider and the factory (exactly as the other two providers already do) and switch the stream send to SendWithAuthRetry. With a nil resolver (API-key users) behavior is unchanged. * fix(errhint): gate classification on provider-origin marker Address CodeRabbit review on #1: agentResponseMsg.err can also carry local failures (a tool's 'permission denied', a 'file does not exist', a config error), and broad substrings would have attached a bogus /provider or /model hint to those. Classify now returns Unknown unless the message carries a provider marker the provider layer always attaches (auth error: / rate limit error: / provider error: / provider request error: / provider stream error:). Local errors never draw a provider hint; provider errors sub-classify as before. Added local-failure test cases proving they stay Unknown. * feat(cli): actionable hint on text-mode exec provider errors zero exec printed provider failures as a bare '[zero] <raw error>' with no next step. Append a one-line hint (reusing the shared errhint classifier) for recognized provider failures — 'run `zero auth`', 'run `zero doctor`', etc. The provider-origin gate means non-provider codes (sandbox_error, mcp_error) and JSON/stream-json output are untouched. * fix(tools): set Result.Truncated for budgeted bash output Address CodeRabbit review on #3: budgetBashOutput recorded truncation in meta["truncated"] but never set the Result.Truncated struct field, so consumers reading the struct (like glob/web_fetch/read_minified do) would miss truncated bash results. Return the bool from budgetBashOutput and set Result.Truncated on every return path. * perf(tui): reuse one LSP manager across prompts runAgentWithOptions built a fresh lsp.NewManager per run and shut it down when the run returned, so gopls (and every other language server) cold-started on the first edit of every turn — 200ms-2s of latency the manager's own design (long-lived, reused servers) exists to avoid. Build one manager per session in newModel (cheap: servers start lazily on first Check) and reuse it across runs; the SelfCorrector still gets a fresh checker wrapping the shared manager each run. Torn down with a short deadline in quit(). Runs fall back to a per-run manager when the session manager is absent (cwd unknown, or a directly-built test model). * perf(tui): accumulate streamingText as []byte (O(1) append) m.streamingText += msg.delta was O(len) per delta -> O(n²) across a long generation (a 10k-token code gen allocates hundreds of MB of intermediate strings and the UI gets progressively laggier). Accumulate into a []byte with append (O(1) amortized) and read via streamingTextString(). A []byte rather than strings.Builder because the TUI model is copied by value on every Update, which would trip strings.Builder's copy check; []byte is nil-safe and copies cleanly like the other slice fields. * perf(tui): coalesce streamed text deltas to one frame Every OnText delta was its own tea.Msg, so a fast provider (100+ tok/s) drove 100+ full Update->View cycles per second, each re-parsing the growing markdown — visible stutter over SSH and wasted CPU. Batch agentTextMsg deltas at the runtime sink over a ~16ms frame and forward them as a single message, decoupling render rate from token rate. Any non-text message flushes pending text first so ordering with tool-call / reasoning / row messages is preserved; the final agentResponseMsg (a tea.Cmd return, not a sink message) is safe because the model already drops deltas for an inactive runID. * fix(agent): raise tool-failure stop threshold 4->6 guardrails.go halted a run after 4 consecutive same-error tool failures. A corrective hint fires at 2, and a model iterating on a genuinely tricky edit can legitimately fail a couple more times after the hint while converging — stopping at 4 cut those runs short. Raise to 6; the streak still resets the instant the tool succeeds or hits a different error, so only true same-error loops are affected. * fix(agent): more reconnect retries with jitter; narrow to transport errors #19: bump maxStreamReconnects 2->4 and add up-to-50% jitter on the exponential backoff (capped at 8s), so a multi-second network blip is ridden out instead of killing the run on a 2s hiccup, and concurrent runs (swarms, cron fleets) don't reconnect in lockstep. #34: drop the '502'/'503' substring matches from shouldReconnect. 503 already exhausted providerio.SendWithRetry (retrying here is a redundant double-retry) and 502 is non-idempotent by providerio's rule (the POST may have been processed). Only genuine transport failures — where no response was received — reconnect now. * fix(agent): reduce stall retries 2->1 to bound stuck-session time A no-output stream stall is detected only after the full stream idle timeout (~5min) elapses, and each retry can idle again — so 2 retries left an interactive session frozen for ~15min. Drop to 1 retry: keeps the common single-hiccup recovery while bounding the worst case to ~2x the idle timeout. * fix(tui): serialize coalescer forwarding to preserve message order Address CodeRabbit review on #7: flush() and the run-switch path drained the text buffer under c.mu but called forward() after releasing it, so a timer-fired text flush could race a concurrent non-text send() and land after the tool-call/reasoning message it should precede. Hold c.mu across drain AND forward (drainAndForwardLocked), so whoever holds the lock delivers atomically and the other caller blocks until it is done — text can never overtake a following non-text message. Added a concurrency stress test (timer racing an inline send) that passes under -race. * feat(tui): accept /compact now Bare /compact already triggers compaction, but /compact now — what users reach for when the context gauge climbs — hit the usage-error branch. Accept the 'now' keyword and advertise it in the command usage. * feat(tui): add /retry, /edit, /copy, /export commands /retry resends the last prompt; /edit recalls it into the composer to tweak and resend — both read a new lastPrompt field captured verbatim (pre-expansion) in launchPrompt. /copy puts the last answer on the clipboard via the existing copy machinery; /export writes a plain role-prefixed transcript to a file (timestamped default, or a given path). Real gaps for SSH users with no mouse select. * feat(tools): separator-insensitive tool_search matching tool_search ranked deferred tools by exact substring, so a model that typed 'webfetch' (dropping the underscore) matched nothing and looped. Add a separator-squashing fallback ('web_fetch' -> 'webfetch') scored below an exact substring match, so precise queries still rank first. * feat(tools): signal ask_user dismissal distinctly from a blank field Empty answers all rendered as '(no answer provided)', so the model couldn't tell a wholesale dismissal (user closed the prompt without answering anything) from one field left blank amid real answers — and might invent a default. FormatAskUserAnswers now flags a full dismissal up front as a skip and marks individual empties '(left blank)' vs '(skipped)'. * feat(tui): fuzzy ranking in model/session pickers The pickers filtered by flat strings.Contains with no ranking, so finding 'sonnet 4.5' among 50+ models meant an exact prefix or scrolling. Rank matches (exact < prefix < contains < subsequence, reusing fuzzySubsequenceGap) so the closest match lands on top, and 'snt45' now matches 'Sonnet 4.5'. Groups stay contiguous — ordered by their best match, never split — so the grouped model picker still renders one header per provider. Covers both /model and /resume (shared commandPicker). * perf(agent): cache per-tool schema render across turns partitionTools re-ran the recursive schema->map conversion (schemaToRuntimeMap) for every tool on every turn — redundant, since a tool's advertised name/description/schema is stable for the run. Add a per-run definition cache keyed by tool name (nil-safe; the plain partitionTools entrypoint and tests pass nil for fresh renders). The partitioning itself (visibility, deferral, ordering) still recomputes every turn — it must, because a tool's deferred state can flip mid-run (swarm tools un-defer once a swarm is active). Only the expensive schema render is memoized. tool_search is excluded by its callers (dynamic description) so it never poisons the cache. * perf(agent): calibrate compaction estimate against real prompt tokens The byte/4 heuristic (ApproxTextTokens) over-counts code-heavy content ~15-20%, so with triggerRatio=0.7 compaction fired at ~60% of true capacity — premature summarizer calls that degrade quality. Each turn now folds (rawEstimate, provider InputTokens) into an EMA calibration ratio, clamped per-sample to a sane band; maybeCompact and the reactive path scale their estimates by it. Turn 1 uses the raw estimate (no data yet); later turns compact near real capacity. * style: gofmt the new agent test files The CI gofmt check flagged the #22/#25 test files (struct-field and switch alignment). No behavior change. * feat(agent): turn-based plan staleness reminder The stale-plan nudge only fired after 10 tool calls since the last update_plan, so a plan that drifts stale across many low-tool-call turns (one call per turn) took many turns to catch — the 'plan set turn 1 stays pending forever' case. Add a turn-count complement: after stalePlanTurnThreshold (8) turns without an update AND with items still pending, fire the same one-shot reminder. Gated on pending items so a fully-completed plan is never nagged; the existing tool-call trigger is unchanged. (The roadmap's fuzzy 'tool calls unrelated to plan items' relevance match is deliberately omitted — too false-positive-prone.) * feat(agent): preserve recent-edits summary across compaction Compaction preserved the plan, loaded skills, tool schemas, and project instructions, but not file edits — so after the editing turns were elided the model knew it had changed a file only vaguely, and would re-read to rediscover its own footprint. Extract write_file/edit_file targets from the elided middle with a one-line note from each tool result, merge them across repeated compactions (newer note per path wins), and carry them in the preserved-state JSON (capped at 20 paths / 160 bytes per note). * feat(tui): workspace-scoped /resume + fewer event reads /resume listed sessions from every project globally, and read the full event file for each of them to test for resumable content. Filter the picker (and /resume latest) by the current workspace Cwd, checked BEFORE the per-session read so a large global history no longer pays N full file reads to build one workspace's list; also skip zero-event sessions via metadata without any read. Sessions with no recorded Cwd (older runs) and an unknown current workspace stay visible so nothing is hidden that can't be confidently placed. Explicit /resume <id> still resolves any session. * feat(tui): warn on model switch that drops staged images Staged images were silently dropped at submit when the active model had no vision support — the user only learned after sending. Chips already render above the composer (renderAttachmentChips); add an immediate amber warning line to the /model switch status (both same-provider and cross-provider paths) when images are staged and the newly selected model can't accept them, so the drop is surfaced at switch time. * feat(tui): inline recovery affordance on setup errors A first-run setup error rendered as a bare red 'error: <text>' with no pointer to the fix. Append a faint, stage-tailored recovery line (edit the endpoint/name/key above then Enter, or pick another model) that names only keys the stage's footer confirms, so the guidance is always accurate. Empty for stages whose footer already makes recovery obvious. * fix: address CodeRabbit review (5 findings) - loop.go: recompute the compaction calibration estimate at calibration time, not request-build time, so a reactive-compaction reissue that shrinks the request feeds the EMA the sample that actually completed. - errhint: match bare HTTP status codes (401/403/429/529) only at digit boundaries, so an incidental number ('4290ms', 'id 14015') isn't mis-bucketed as auth/rate-limit. - bash: drop the stray unmatched ']' in the output-truncation marker. - tui /retry: apply the same exiting/compactInFlight guards commandPrompt has, so a retry can't race compactResultMsg's rewrite of session state. - tui /export: write the transcript 0o600 (it may contain echoed secrets), not world/group-readable 0o644. * test(tui): skip export file-mode assertion on Windows os.WriteFile ignores Unix permission bits on Windows and Stat reports 0666, so the 0o600 assertion failed there. The production 0o600 is honored on Linux/macOS (where it matters); guard the check with runtime.GOOS so Windows smoke passes. * fix(tui): align /resume latest filters + case-insensitive workspace match on Windows Address CodeRabbit review on #21: - latestResumableInWorkspace now applies the SAME filters as the picker (skip zero-event metadata and empty/failed runs), so /resume latest can't land on a session the picker intentionally hides. - sessionMatchesWorkspace compares paths case-insensitively on Windows (its filesystem is), so the same workspace spelled with different casing no longer hides resumable sessions; other platforms stay case-sensitive. * fix(tui): /retry resends the last prompt's image/PDF attachments /retry relaunched only lastPrompt's text, but launchPrompt clears the pending image/document queues once a turn is sent. A vision- or PDF-backed prompt that failed would silently retry as text-only and answer a different task. Snapshot the consumed attachments at launch and re-stage them on /retry so launchPrompt rebuilds an identical request (document preamble, images, and the submit-time vision re-check). startNewSession drops the snapshot so a post-/new /retry cannot leak the previous session's attachments. Also correct the OAuthResolver factory comment, which omitted the Gemini provider it is now passed to. * fix: address jatmn review (reconnect 5xx, /edit attachments, edit-cap, coalescer test) - reconnect: exclude HTTP 5xx (500/502/503/504) before the transport substring match so "504 Gateway Timeout" no longer slips through on the generic "timeout" needle. A gateway timeout is non-idempotent — the completion POST may already have reached the model — so replaying the connect risked duplicate billable work. Reuses a digit-boundary status matcher exported from errhint (HasStatusCode). Test asserts 504/500 stay non-reconnectable. - tui /edit: re-stage the remembered image/PDF snapshot alongside the recalled prompt text, mirroring the /retry fix — editing a vision- or document-backed prompt no longer silently resends a text-only version. - compaction recent-edits: order edits by LAST edit, not first, and move a re-touched path to the newest position on merge, so the maxRecentEdits tail cap keeps the file the model most recently edited instead of dropping it. Adds mergeRecentEdits (edit-specific) and a capped-list regression. - coalescer test: inject the frame timer via afterFunc so TestCoalescerBatchesDeltas proves buffering deterministically instead of racing the real 16ms wall-clock timer. * fix(tools): bound bash output capture in memory, not just model-visible text Previously stdout/stderr streamed into unbounded bytes.Buffers and were only truncated to the 96 KiB budget AFTER the command finished, so a runaway command (cat huge.log, yes) could grow Zero's memory until it stalled or OOMed before budgetBashOutput ever ran. Replace the capture with boundedBuffer, an io.Writer that keeps only the head+tail each stream will surface and discards the middle as it arrives, while counting the true total for the truncation marker. Memory is now bounded to ~head+2×tail per stream regardless of output size. truncateHeadTail is refactored onto a total-aware core so the string path (and its tests) are unchanged, and budgetBashCapture reports the true raw_bytes even though only a bounded slice was held. Adds boundedBuffer/capture unit tests and an end-to-end test that streams ~5× the budget and asserts bounded emission with an accurate raw_bytes. |
||
|
Anandan
|
f401c66e81 |
polish: cut per-turn overhead, stabilize cache, add /new (#478)
* polish: cut per-turn overhead, stabilize cache, add /new Context economy - Compaction: preserveLast 8->6, triggerRatio 0.8->0.7 (both copies in sync). - System prompt: kill the "ELABORATE summary" directive; scale summary to the work. Comment discipline moved into the core prompt for all model families (was Anthropic-only); Minimax/DeepSeek/Qwen/etc. now get it. - Token-budget regression test ratchets the fixed per-turn overhead (base prompt + eager tool schemas) using the runtime's own estimator, so silent prompt/schema growth fails CI. Response quality - defaultResponseStyle balanced->concise. Reversible via /style. Tool discovery - <available_skills> block in the system prompt lists installed skills (name + one-line description) so the model loads the right one on the first call instead of guessing and reading the failure. Budget-capped at 640 bytes with an "...and N more" fallback; renders nothing when no skills are installed (byte-identical to prior prompt). Session lifecycle - /new starts a fresh session in place; previous session stays on disk and is named in the confirmation for /resume. Guards against mid-run reset. - /clear now notes that context persists and points to /new. Performance & caching - partitionTools no longer alpha-sorts the whole tools array every turn, which inserted a loaded deferred tool mid-array and shifted the cached prefix on every load. It now builds append-only regions [eager sorted] + [tool_search] + [loaded deferred sorted] so a mid-session load grows the tail instead of busting the eager prefix. Regression test asserts the eager block precedes any loaded tool. - /context now shows cache hit rate (flags low rate over >5 turns as a possible unstable prefix) and rolling average turn latency with time-to-first-token. Both reset on /new. Measured baselines (auto-mode, no workspace): base prompt 3162 tokens, eager tool schemas 2433 across 12 tools. Budget ceilings set ~10% above. Deliberately deferred (need a decision): inverted safety-policy rewrite (sandbox doesn't catch ~half the B-tier rules), core-tool/web_fetch deferral (activation-semantics blocker), aggressive prompt slim. * fix(tui): clear staged attachments + queued message on /new CodeRabbit: startNewSession reset conversation/usage/latency state but not the staged input (pendingImages, pendingImageLabels, pendingDocuments, queuedMessage), which is only consumed at prompt-submit. Without clearing it, the fresh session's first prompt silently inherited the previous session's images/documents/queued text. Test now stages all four and asserts they clear. |
||
|
euxaristiaandClaude Fable 5
|
e4f760ee8b |
fix(agent): reject a malformed additional_permissions payload before prompting (#453)
* fix(agent): reject a malformed additional_permissions payload before prompting A shell tool call that sets sandbox_permissions: with_additional_permissions but omits (or malforms) the additional_permissions payload was accepted into the normal permission-prompt flow. buildPermissionEvent silently swallowed the validation error while rendering the prompt's scope text, so the user saw an ordinary-looking request, approved it, and only then hit a hard denial from the same validation running again at grant time (loop.go's grantInlineAdditionalPermissions), with the confusing message "failed to grant session additional permissions: missing additional_permissions". A malformed payload can never succeed regardless of the user's decision, so run the same check (inlineAdditionalPermissionsProfile, already the single source of truth for this shape) up front, before any prompt, and return a plain tool error the driving model can see and retry from. This also covers the mismatched case (additional_permissions present without the flag). Co-Authored-By: Claude Fable 5 <[email protected]> * fix(agent): make the additional_permissions rejection message actionable The rejection added in the previous commit correctly stops a malformed sandbox_permissions: with_additional_permissions call before any prompt, but its message ("missing additional_permissions; provide at least one of network or file_system") gave a model nothing concrete to correct itself with. Observed in practice: a weaker tool-calling model (xai/grok-build-0.1) kept re-attaching sandbox_permissions out of habit from an earlier, correct gh/git call onto later commands that didn't need it (network was already open via a session-scoped grant), hit this same rejection 4 times in a row with no change in its retry, and gave up. Both messages (missing payload, and empty-but-present payload) now include a concrete, schema-accurate example - {"network": {"enabled": true}} or {"file_system": {"write": ["/path"]}} - and the missing-payload message adds the other half of the fix: if the command doesn't actually need elevated permissions, omit sandbox_permissions entirely and retry. Co-Authored-By: Claude Fable 5 <[email protected]> --------- Co-authored-by: Claude Fable 5 <[email protected]> |
||
|
KRATOS
|
952788f72d |
fix(tui): resolve every permission request so the agent can't deadlock (#397)
A goroutine dump of a run "stuck for 33 minutes" showed the agent parked forever in executeToolCall -> requestPermission, blocked on the TUI's decision channel. Root cause: a shell command whose sandbox decision was Allow but which requested additional sandbox permissions (sandbox_permissions: with_additional_permissions) made shouldRequestPermission block on OnPermissionRequest, yet buildPermissionEvent stamped the event Action=allow. The TUI's permissionRequestMsg handler only rendered/resolved Action=prompt requests and silently dropped everything else — so the decide callback was never called, the agent waited on a channel nothing would signal (no ctx cancellation on the active run), and the UI showed "Working" with no prompt and zero network activity. Intermittent and model-dependent: gpt-5.5's folder-creation command set the additional-permissions flag, glm's did not, which is exactly why one hung and the other ran clean. Two-part fix: - agent (buildPermissionEvent): an additional-permissions request is an ELEVATION the user must consent to, even over a sandbox Allow, so mark it Action=prompt (matching shouldRequestPermission's own condition). It now surfaces as a real prompt instead of a dropped auto-allow. - tui (permissionRequestMsg handler): enforce the contract that a blocked agent ALWAYS gets a decision. A non-prompt action resolves immediately and FAILS CLOSED (deny; cancel stays cancel) so the UI can never silently grant an unsurfaced permission; a superseded-run (runID mismatch) resolves as cancel so its parked goroutine unblocks without depending on ctx-cancel ordering. Matches the ACP handler's always-resolve, fail-closed contract. Red-green: reverting the tui branch reproduces the deadlock (decide never called); reverting the agent override reproduces Action=allow. End-to-end against a mock backend emitting the exact triggering tool call, the real TUI now renders the permission prompt and the run completes on approval, where before it parked forever. |
||
|
KRATOS
|
02f0c0929b |
fix: cut the content-stall dead-wait and auto-retry a mid-tool-call stall (#362)
* fix: cut the content-stall dead-wait and auto-retry a mid-tool-call stall Root-caused (via a verified multi-agent audit) the chatgpt/gpt-5.5 "stuck" reports: the ChatGPT Codex backend pauses real generation mid-stream on large single tool-call outputs (a ~1000-line write_file) while continuing SSE keep-alive heartbeats. That is an upstream limitation, not a Zero bug or a connection-pooling issue (the darwin DisableKeepAlives fix is applied and confirmed not the cause). Zero already detected it, but two UX gaps made it feel like a hang: 1. Faster abort. The content-stall watchdog fired at idle*2 = 10 minutes. A heartbeat-pause stall on gpt-5.x rarely recovers, so 10 minutes of a frozen token counter was a terrible wait for a doomed turn. Replaced the StreamContentStallFactor=2 const with ContentStallTimeout(idle)=idle*1.2 (6 minutes at the default). It still only ever fires on a TRUE zero-output freeze — any real data line resets it — so healthy slow-reasoning runs are unaffected. Scales with ZERO_STREAM_IDLE_TIMEOUT; returns 0 (disabled) when idle is disabled. 2. Auto-retry the actual stall. The existing stall-retry only fired when NOTHING had been forwarded — so the common case (model streamed transient reasoning + began the big write_file, then froze mid-arguments) was NOT retried, because the incomplete tool call is collected and transient output was forwarded. Relaxed the gate to retry when no visible PROSE was shown (forwardedVisibleText, marked only by OnText) and no final text was collected — dropping the len(ToolCalls)==0 exclusion, since an incomplete tool call from a timed-out stream is never executed (the turn returns on the error before dispatch) and never appended to messages, so the retry re-sends clean context with no duplication. A visible "[no output — model stalled; retrying N/max]" notice fires per attempt. Forwarded prose still blocks the retry (it would duplicate answer text). Verified: full -race suite green; a 4-dimension root-cause audit and a 3-dimension adversarial review of this diff both passed with no confirmed defects. New tests: TestRunRetriesStallAfterIncompleteToolCall (the exact gpt-5.x case), TestRunRetriesStallAfterOnlyReasoning; preserved TestRunDoesNotRetryStallAfterPartialOutput (prose still blocks retry). * fix(providerio): make ContentStallTimeout overflow-safe Address PR #362 review: idleTimeout*6/5 could overflow int64 for a pathologically large ZERO_STREAM_IDLE_TIMEOUT and wrap to a NEGATIVE duration, which would arm the content-stall timer to fire immediately and abort every stream. Compute 1.2× as idle + idle/5 with a clamp to the max duration on overflow (which just means "effectively no content watchdog", the sane result for an absurd idle timeout). The normal 1.2× values are unchanged (5m->6m, 30s->36s, 100ms->120ms). New test: TestContentStallTimeout covers the scaled values, the zero/disabled case, and the MaxInt64 overflow guard. |
||
|
KRATOS
|
7fb0092de7 |
fix: agent runtime resilience — stream stalls, weak-model tool args, and sub-agent reliability (#349)
* fix(local): defeat the macOS stale-pooled-connection hang (transport + stall-retry)
Root cause (you confirmed it: both chatgpt AND ollama hung, only on macOS): providers
fell back to http.DefaultClient, whose pooled keep-alive connections get reused on a
later turn after the server/NAT silently dropped them. The model call is a POST
(non-idempotent), so Go won't auto-retry it on a fresh conn — it blocks forever.
macOS keeps dead pooled conns around longer than Linux, so it reproduced on macOS only.
1. providerio.HTTPClient now returns a shared, tuned transport (ResponseHeaderTimeout
60s bounds a hung POST that gets no response; IdleConnTimeout 30s stops dead conns
from lingering to be reused). Routed openai/codex through it too (anthropic/gemini
already used HTTPClient). Slow first tokens / long reasoning are unaffected — the
header timeout only bounds time-to-first-response-header, not the stream body.
2. Agent loop now retries a stream idle/stall timeout that produced NO output yet, on
a fresh connection (capped + backoff). A partial-output turn is NOT retried (would
duplicate). Recovers any stall that slips past the transport timeout.
Local only — not pushed.
* fix(local): tolerate concatenated tool-arg JSON + widen response-header timeout
1) Weak models (minimax-m3, glm) pack MULTIPLE concatenated top-level JSON objects
into one tool-call arguments slot ({A}{B}), which json.Unmarshal rejects with
'invalid character {' after top-level value' (~9 failed tool calls/session).
New recoverableToolArguments returns the first JSON value ONLY when the payload is
one value optionally followed by more WHOLE JSON values; it rejects a bad/truncated
first value OR trailing NON-JSON garbage ({"x":1}xyz) so genuine corruption still
errors (adversarial-review fix). decodeToolArguments uses it at the central dispatch
(executeToolCall) so {A}{B} runs the first object; historySafeToolCalls uses the
SAME helper so the replayed transcript records the recovered first object (not {}),
keeping dispatch and history consistent. Truly-malformed JSON still hits the
existing sanitize-to-{} + retry path. Trailing objects are dropped (the model
re-issues on a later turn).
2) Raise the shared transport ResponseHeaderTimeout 60s -> 120s so a slow cloud proxy
(ollama *:cloud) that withholds its 200 header until first token isn't wrongly
aborted; still bounds a truly dead reused connection.
Local only — not pushed.
* feat(local): add /turns command + raise default maxTurns 30->50
Agentic multi-step runs (build + delegate + swarm + MCP + self-verify) routinely hit
the 30-turn ceiling mid-task and never reached the later steps; removing /mode earlier
also left NO interactive way to raise the budget.
- /turns [n]: per-session tool-turn budget setter (commandGroupSession), mirroring
/selfcorrect. Empty/status shows the current budget; a positive int sets
agentOptions.MaxTurns (read by the run at model.go); clamped to maxTurnsCeiling=500
to guard against typos; invalid/non-positive shows usage without changing it.
- defaultMaxTurns 30 -> 50 (matches the old 'deep' preset). Raise per-run with /turns
for long tasks (e.g. /turns 150 for a full acceptance run).
Local only — not pushed.
* fix(local): swarm reports failed members as failed + sub-agents inherit /turns budget
Issue B — a swarm member whose child exited non-zero (e.g. exit 4 / max-turns) was
recorded [done] (swarm_status '0 failed'), so the orchestrator — which can't see the
AGENTS panel — trusted incomplete work. The specialist launcher ignored
res.Result.Status; it now surfaces a StatusError result as a member FAILURE
(errors.New(report)) so the swarm marks it [failed]. Added Coordinator.FailWithSession
and switched the member watch path to it so a failed member keeps its session id
(still drillable) and its report rides along as the failure message.
Issue A — a member hit its OWN turn limit because /turns raised only the parent's
budget; children re-resolved config.json's default. New config.SetMaxTurnsEnv exports
ZERO_MAX_TURNS, read in applyEnv -> cfg.MaxTurns; /turns now exports it so spawned
sub-agents / swarm members (which inherit the env) run with the same budget. A
non-numeric/zero env value never clobbers the configured budget.
Tests: launcher non-zero-exit -> failed (+ session kept); SetMaxTurnsEnv round-trip
+ garbage-env guard; /turns exports + clamps the env. Full gate green (73 pkg).
Local only — not pushed.
* fix(local): corrective error for unknown specialist name + clamp ZERO_MAX_TURNS
1) A model that invents a specialist name (validator-runner, file-creator,
file-writer — none registered) got an opaque 'specialist not found' and kept
spawning doomed sub-agents. loadManifest now lists the available specialists and
notes that 'worker' has shell/edit while explorer/code-review are read-only, so the
model self-corrects to a capable specialist. New availableSpecialistList helper.
2) Defense-in-depth (review nit): applyEnv now clamps an inherited ZERO_MAX_TURNS to
MaxTurnsCeiling, so a raw shell 'export ZERO_MAX_TURNS=999999' can't bypass the
/turns ceiling. Ceiling promoted to config.MaxTurnsCeiling (single source of truth,
referenced by the TUI /turns clamp).
Tests: unknown-name error lists worker/explorer/code-review; applyEnv clamps an
over-ceiling env value. Full gate green (73 pkg).
Local only — not pushed.
* fix(local): surface sub-agent kill signal instead of opaque exit -1
When a sub-agent child is terminated by a signal, command.Wait returns an ExitError
whose ExitCode() is -1, losing the cause. Capture ProcessState.String() (portable,
e.g. "signal: killed") into ChildRunResult.Signal and have BuildFinalResult render
it — "Subagent terminated by a signal (signal: killed) — killed before it finished.
Likely causes: the OS out-of-memory killer ... a timeout, or cancellation." — instead
of an opaque "Subagent failed (exit -1)". The message lists causes rather than
asserting OOM, since timeout/cancellation also SIGKILL the child. A captured signal
always surfaces even if a late run_end reported a clean exit. The background task
path (which can't carry the signal name) emits the same hint for a negative exit code.
* address CodeRabbit review on #349
- loop.go: gate stall-retry on a turn-level forwardedAnything flag (wrapped
callbacks), so a retry never re-streams on top of output already forwarded earlier
in the turn (e.g. across the reactive-compaction retry). New test covers a
reasoning-then-stall (collected.Text empty but output forwarded).
- loop.go: route a reissued stream's non-stall error through the SAME recovery as
the initial stream (image-rejection wrapping / context-limit compaction) via a
shared recoverStreamError closure, instead of returning it raw.
- exec.go: drop the hardcoded worker/explorer/code-review guidance from the
unknown-specialist error; rely on the dynamically-rendered available list (a
custom registry may not have those names).
- streamer.go: kill-signal message lists OOM/timeout/cancellation evenhandedly
instead of asserting OOM (the branch also covers timeouts/cancellations).
- launcher_specialist.go + exec.go: preserve the child SessionID on a post-start
failure too, so FailWithSession keeps those failed members drillable.
- model.go: block /turns while a run is active (mid-run change would make the
inherited ZERO_MAX_TURNS budget inconsistent for later-spawned sub-agents).
|
||
|
KRATOS
|
fb3277f46e |
feat(tui): tabbed ask_user questionnaire in the composer with suggested answers (#327)
* feat(tui): tabbed ask_user questionnaire rendered in the composer
Redesign the interactive ask_user prompt to match the requested mockup: when a
question is raised the composer text box becomes the questionnaire.
- Multi-question prompts render as a row of TABS (one per question + a trailing
Confirm tab). Tab / Shift+Tab switch questions, answers are given in any order,
and Enter on Confirm submits them all. A single-question prompt submits on
answer (no Confirm step) and shows no tab row.
- Each option can carry a one-line DESCRIPTION (contract: per-option
optionDescriptions, or option objects {label, description}); options render
numbered with the description dimmer underneath and the recommended one tagged.
- Per-question optional `header` gives each tab a short title (falls back to a
trimmed question).
- Rendered in the composer/footer region (footerView), replacing the text box
while active, instead of a scrolling transcript card.
- Keys: Tab move · ↑↓ select · Enter confirm · Esc dismiss; a printable key in
the picker drops into free-text; Esc from "type your own" returns to the picker;
multi-select questions are free-text with the options shown as suggestions.
- Headless unaffected (zero exec never wires OnAskUser).
Removed the old single-question sequential state (cursor/typing/index,
renderFocusedAskUserPrompt, resolveAskUser). Tests rewritten for the tabbed flow
(picker default, type-your-own, multi-question tabs + Confirm, Esc dismiss,
descriptions render). make build / go vet / go test ./... -race / make lint green.
* fix(tui): ask_user questionnaire renders on the black terminal canvas
It was painting a gray panel background (styledBlockFill with zeroTheme.panel +
onPanel-wrapped foregrounds). Since the questionnaire replaces the composer, it
should match it: bare foregrounds on the terminal canvas (black) with the
composer's lineStrong border and no fill. The lime cursor/badge highlights and
the recommended tag stay.
* fix(tui): address PR review — schema/parser alignment, single-question Tab
- ask_user schema: options.items is declared as string, so advertise only the
string form (descriptions go via the parallel optionDescriptions array). The
parser stays object-tolerant defensively, but the schema no longer tells
strict-schema providers to send objects it would reject.
- moveAskUserTab no-ops for single-question prompts (no tab strip / Confirm tab),
so Tab/Shift+Tab can't move the prompt into the hidden Confirm state. Test added.
|
||
|
KRATOS
|
1d1fa7c514 |
feat(tui): ask_user suggested answers with a recommended default (#326)
* feat(tui): suggested answers with a recommended default for ask_user When the agent asks a clarifying question mid-task, it can now offer 2-4 suggested answers and mark one recommended; the interactive TUI renders a selector (arrow keys, lime cursor, "(recommended)" tag) with the cursor resting on the recommended option, plus a trailing "✎ Type my own answer…" entry that drops into the existing free-text composer. Whichever the user picks — a suggested option's text or their typed answer — flows through the same answer channel the agent loop already waits on. Optional and backward-compatible: a question with no options renders the plain free-text prompt exactly as before. Headless `zero exec` is unaffected (it never wires OnAskUser; the non-interactive best-assumption fallback stands). - contract: agent.AskUserQuestion / tools.AskUserQuestion gain Recommended; the tool schema advertises optional options + recommended; parser keeps recommended only when it resolves to one of the options (canonical text). - system prompt: tells the model it MAY supply options + a recommended one. - TUI: pendingAskUserPrompt gains cursor/typing; ask_user_prompt.go holds the selector logic mirroring permission_prompt.go; renderFocusedAskUserPrompt renders the picker or the free-text prompt. Tests: selector defaults to recommended and returns its text; "type my own" falls back to free-text and returns the typed text; no-options stays plain free-text. make build / go vet / go test ./... -race / make lint green. * fix(tui): ask_user — single input, Esc steps back from "type my own" Two UX fixes from interactive testing: - No double display: the focused ask card no longer echoes the in-progress free-text answer (it was shown both inside the card AND in the composer text box). The composer is now the single input; the card just shows the question and a hint. renderFocusedAskUserPrompt no longer takes the input string. - Esc from "type my own" returns to the selector for that question instead of cancelling the whole questionnaire (escapeAskUser); a question with no options, or the selector itself, still cancels on Esc as before. Tests: type-my-own asserts the card does not echo the typed text and the hint flips to "type your own answer"; new test covers Esc returning to the selector (no cancel, no answer delivered) then selecting a suggested option. * fix(tui): ask_user review fixes — multi-select, selector typing, dead code Address the three confirmed findings from the code review: - multi-select no longer silently narrowed to one option: a single-pick selector cannot represent MultiSelect, so a multi-select question now renders as free-text with its options surfaced as suggestions; the user types one or more and the answer is returned verbatim (syncQuestionState; escapeAskUser and the renderer updated to match). - typing in selector mode no longer discarded: a printable keystroke while the selector is showing flips into free-text and captures the key (model.go key capture), instead of accumulating in a hidden input that Enter then overrode with the highlighted option. - removed dead submitAskUserAnswer (the Enter handler now routes through confirmAskUser; the orphan had no callers). Tests: multi-select falls back to free-text and returns the typed answer; a printable key in the selector switches to free-text and is captured. Full gate green (build / vet / go test ./... -race / make lint). * fix(tui): address PR review — element-wise option compare, clearer Esc copy - ask_user_test: compare option slices with slices.Equal instead of strings.Join, so the parser test can't be fooled by ["A,B"] vs ["A","B"] or nil vs [""]. - ask_user prompt: the Esc hint said "skip", which reads like "skip this question" on a multi-question prompt even though Esc ends the rest of the questionnaire (remaining answered by best-assumption). Reworded to "skip the rest" in the selector and free-text hints; "type my own" keeps "go back". |
||
|
Anandan
|
b8b0bb7f65 |
Move deferred tool discovery into tool search (#324)
* Move deferred tool discovery into tool search Keep deferred tool discovery in the tool_search definition instead of appending it as a per-turn user message, so models do not answer or acknowledge internal discovery metadata. Preserve deferred loading behavior: hidden tools remain hidden until loaded, tool_search stays reachable for allowlisted deferred tools, and retry paths keep loaded/deferred state. Tested: env GOCACHE=/tmp/zero-go-cache go test ./internal/agent ./internal/tools -run 'Test.*Deferred|Test.*ToolSearch|TestRunLoadsDeferredToolThenAdvertisesNextTurn|TestRunReactiveRetryKeepsLoadedDeferredToolAndDiscovery|TestRunConnectTimeReactiveRetryKeepsLoadedDeferredToolAndDiscovery|TestAllowlistedDeferredToolsKeepToolSearchReachable|TestDisabledToolSearchFallsBackToEager' Tested: env GOCACHE=/tmp/zero-go-cache go test ./internal/tools Tested: env GOCACHE=/tmp/zero-go-cache go test ./internal/agent ./internal/tools ./internal/cli Tested: git diff --check * Clarify deferred tool search selection |
||
|
KRATOS
|
f8ac5eceb0 |
fix(agent): gate headless run completion on honesty (no false-success on no-tool-call / self-reported-incomplete turns) (#325)
* fix(agent): don't end a run as success on a no-tool-call turn mid-task
A turn that produced text but no tool call was always accepted as the final answer, so the loop reported success even when the model stopped mid-task (e.g. ended on "...Let me check the SSH configuration:" with plan steps still pending).
Add an opt-in completion gate (Options.RequireCompletionSignal): when a turn has no tool call and work clearly remains -- pending update_plan items, or the message ends on a continuation cue -- re-prompt the model to continue instead of finalizing. Bounded by maxContinueNudges and still by MaxTurns/the deadline; once the budget is spent the run finalizes as INCOMPLETE (Result.Incomplete) rather than success. Default off, so the interactive path is byte-identical.
Genuine single-turn completions (no pending plan, no cue) still finalize as success. Covered by internal/agent/completion_gate_test.go.
* feat(exec): report stalled headless runs as INCOMPLETE (exit 4)
Enable the agent completion gate for headless exec (RequireCompletionSignal) and map Result.Incomplete to run_end status "incomplete" with a new exit code 4, so a run that stalled mid-task (model stopped without a tool call while work remained, continue budget exhausted) is no longer reported as success. Interactive callers are unaffected.
* fix(agent): downgrade self-reported non-completion; advisory task-grounded acceptance
Reduce -- not eliminate -- false-success on headless runs. Two DETERMINISTIC, unit-tested gates (with the plan gate from the prior commit):
(a) self-report downgrade: if the final message admits the model guessed or could not meet the objective, finalize INCOMPLETE (exit 4), never success. Inability is matched by first-person STEMS generalized over verb/tense ("I cannot/can't/could not/am unable to/do not have/unable to ...") plus guess/fallback/uncertainty phrases, with a guard so success-y negations ("could not find any issues", "cannot reproduce") are not misread. (b7bc0b8's plan gate already forces INCOMPLETE on pending/in_progress update_plan items at termination.)
(b) task-grounded acceptance is ADVISORY, not a guarantee. When --self-correct is on it demands one bounded acceptance pass that re-derives the task's stated criterion and runs a concrete check, discouraging three false-success patterns (well-formed==correct, existing-tests-pass==objective-met, result==baseline-it-was-told-to-beat). But it is a prompt: a model that ignores it and confidently claims "PASS, all requirements met" still slips, because ZERO has no general oracle to verify correctness against a task's hidden criterion. Empirically (TB-2, qwen3-coder:480b) this reliably catches admissions and incomplete plans and REDUCES false-success, but a confident false PASS on a model-ceiling task is a residual, fundamental gap -- not a tuning miss.
Default off (RequireCompletionSignal); interactive callers are byte-identical. Covered by internal/agent/{acceptance_gate_test.go,completion_gate_test.go}.
* feat(exec): surface the INCOMPLETE reason in run_end and logs
When a headless run finalizes as INCOMPLETE, include Result.IncompleteReason in the session error event and a stderr warning so an honestly-incomplete run (e.g. "the final message admits the objective was not met") is debuggable rather than an opaque exit 4.
* fix(agent): harden completion gate per PR review
Addresses the correctness findings from the PR review.
self-report (#1): drop behavior-describing phrases ("fall back to",
"placeholder value", bare "best guess", "as a fallback", "without proper")
that also match legitimate final answers; keep first-person/uncertainty
admissions only, since these are matched without a context guard.
self-report (#5): scan every occurrence of each inability stem so an early
success-negation ("could not find any examples") no longer masks a later
genuine admission with the same stem ("could not implement it").
continuation cue (#6): require a trailing colon AND an action lead-in on the
final clause; stop flagging recommendations, plain summary colons, and
sign-offs. Still catches the mid-line "...Let me check the config:".
gate order (#8): check the self-report admission BEFORE pending-plan, so an
admitted-impossible task downgrades immediately with the accurate reason
instead of burning continue-nudges.
pending plan (#3): treat a pending/in_progress update_plan item as a
NUDGE-only weak signal -- it no longer forces INCOMPLETE on its own (a
completed run that left stale plan bookkeeping is trusted). Only a
continuation cue or a self-report admission finalizes INCOMPLETE.
max-turns (#4): a run cut off at the MaxTurns ceiling now finalizes
INCOMPLETE under the gate instead of being reported as success.
exec json/cron (#2, #7): for -o json, emit the terminal done with exit 4 on
an incomplete run (final() pre-emits a success done:0 for json that would
otherwise mask it); emit an error event -- not just a warning -- so the cron
failure extractor can recover the reason.
Deferred (noted on the PR): acceptance-only-when-mutated (#9, cost) and
reusing tools.normalizePlanStatus (#10, would widen scope to internal/tools).
Tests: add TestContinuationCueMatching and TestMaxTurnsCutoffIsIncompleteUnderGate,
extend TestSelfReportedIncompletionMatching with the #1/#5 cases, and replace
the in_progress=>incomplete test with TestPendingPlanAloneDoesNotForceIncomplete.
make build / go vet / make lint / go test ./... -race all green.
|
||
|
Anandan
|
d29ab98a79 |
Improve sandbox permission handling (#323)
Retry shell commands outside the sandbox when sandboxed output only reflects the sandbox namespace, and keep network retries sandboxed when a network grant is enough. Show user-facing permission reasons for escalated shell commands instead of exposing internal sandbox policy text. Tested: env GOCACHE=/tmp/zero-go-cache go test ./internal/agent ./internal/tui -run 'TestRunUsesEscalatedJustificationForPermissionPrompt|TestRunUsesUserFacingEscalatedFallbackForPermissionPrompt|TestPermissionPromptMapsEscalatedSandboxReason|TestRunRetriesShellUnsandboxedAfterSandboxNamespaceLimitedOutput' Tested: env GOCACHE=/tmp/zero-go-cache go test ./internal/tools Tested: env HOME=/tmp/zero-test-home-clean XDG_CONFIG_HOME=/tmp/zero-test-config-clean GOCACHE=/tmp/zero-go-cache go test ./... |
||
|
Anandan
|
80ab238361 |
Align sandbox command approval behavior (#322)
* Align shell approval sandbox execution Make approved shell command-prefix executions use the escalated sandbox path when denied-read policy allows it, so remembered approvals do not still hit the restrictive sandbox. Allow auto sandbox mode to degrade to a direct command plan when native command wrapping is unavailable, while strict sandbox require continues to fail closed. Update regression tests for degraded execution, prefix-approved shell calls, and unavailable native sandbox metadata. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent ./internal/sandbox Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/tools -run 'TestRegistryRunsWithDegradedUnavailableNativeSandbox|TestBashToolRunsWithDegradedUnavailableNativeSandbox|TestBashRequireEscalatedBypassesNativeSandboxAfterApproval|TestExecCommandRequireEscalatedBypassesNativeSandboxAfterApproval' Tested: GOCACHE=/tmp/zero-go-cache go test ./... -run '^$' Tested: git diff --check * Align sandbox command approval behavior Evaluate reusable shell approvals across plain command segments so prefix grants can cover commands with safe read-only tails without relying on command-specific sandbox exceptions. Keep process inspection commands sandboxed by default; unsandboxed host access continues to require explicit escalation or a narrow command-prefix approval. Improve shell tool guidance around sandboxed additional permissions, full escalation, justifications, and narrow reusable prefixes. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox ./internal/tools ./internal/agent Tested: env HOME=/tmp/zero-test-home-clean XDG_CONFIG_HOME=/tmp/zero-test-config-clean GOCACHE=/tmp/zero-go-cache go test ./... * Normalize explicit sandbox backend platforms Infer the platform from a caller-provided sandbox backend name before choosing the manager runtime platform. This keeps Linux, macOS, and Windows backend plans deterministic across host OSes when tests or callers provide a backend without an explicit Platform field. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox ./internal/tools Tested: env HOME=/tmp/zero-test-home-clean XDG_CONFIG_HOME=/tmp/zero-test-config-clean GOCACHE=/tmp/zero-go-cache go test ./... |
||
|
Anandan
|
f1502f0bfe |
Add local control browser and terminal tools (#319)
* Add local control browser and terminal tools Add configurable local browser, desktop, and terminal control helpers with packaged helper discovery for install and release flows. Improve sandbox scope classification for local control commands and shared temporary paths. Add structured browser actions, terminal session line input/key normalization/read support, and artifact capture for local-control outputs. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/localcontrol ./internal/tools ./internal/cli ./internal/config -run 'Test.*LocalControl|TestTerminal|TestBrowser|TestDesktop|TestCapture' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/localcontrol ./internal/tools ./internal/cli ./internal/config ./internal/agent Tested: tuistory smoke launched bash, sent interactive input, waited for expected output, and captured a no-cursor snapshot Tested: git diff --cached --check * Fix npm wrapper helper manifest test on macOS Canonicalize expected helper paths in the npm wrapper manifest test so macOS /var and /private/var temp path aliases compare correctly. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/npmwrapper Tested: git diff --check * Harden local control review fixes Respect resolved local-control config before tool listing and filter validation while preserving stream-json usage error ordering. Harden browser, desktop, and artifact tool validation; guard Linux-only app launch behavior; clear stale helper manifests from the npm wrapper. Make helper packaging lockfile-based with npm ci and add installer/package tests for helper staging. Tested: focused local-control/tool/CLI/release/install/sandbox suite Tested: isolated full go test ./... Tested: macOS and Windows compile-only checks for touched packages |
||
|
Vasanth T
|
46c44a8753 |
feat(swarm): let swarm members build (write + sandboxed shell) (#313)
* feat(swarm): let swarm members build (write + sandboxed shell) Swarm subagents were effectively read-only: a member runs headless at "--auto low" (PermissionModeAuto), and ToolAdvertised strips every prompt-requiring tool from the model's list — so write_file/edit_file/ apply_patch/bash were never handed to the member. It only saw read/grep/ glob/plan tools, so it could not build and the orchestrator did all the writes itself. Add PermissionModeMemberAuto: a headless mode that ADVERTISES the in-workspace mutators (SideEffectWrite + SideEffectShell) on top of the Auto set, while behaving exactly like Auto everywhere else. The sandbox engine still decides at call time — in-workspace writes and sandbox-backed shell auto-allow; out-of-workspace writes, network, and destructive commands prompt and so are denied headless. member-auto normalizes to Auto in the sandbox layer, so no sandbox authority is widened beyond what an interactive auto agent already has. Scope: swarm members only. The launcher sets MemberAutonomy; the executor emits "--auto member" for a non-unsafe member (unsafe still -> high, plain specialists still -> low). Task-tool specialists are unchanged. Windows note: shell only auto-runs when the OS sandbox is active; without `zero sandbox setup` a member writes files but cannot run builds. * style: gofmt member_auto_test.go |
||
|
Anandan
|
6be50cd3a7 |
fix(sandbox): retry approved network shell commands (#301)
* fix(sandbox): retry approved network shell commands * test(sandbox): fix portable smoke expectations * fix(sandbox): restore Windows offline egress block |
||
|
KRATOS
|
63c0e278ff |
feat(tui): chat + context sidebar redesign, AGENTS panel, animations (#300)
* feat(tui): chat + context sidebar redesign, AGENTS panel, animations, plan-timer fix
Two-column managed-mode layout: the existing chat (scroll engine, reduced width)
on the left + a new right context sidebar — AGENTS (spawned subagents: status
glyph, name, and a live working detail for running ones, with a running/total
count), the live PLAN, and a token readout. The sidebar is suppressed on the
home screen and under overlays/subchat; alt-screen managed mode, wide terminals.
Also: freeze the plan clock while the agent is idle (an in_progress step left
mid-plan when the agent yields stops ticking instead of counting forever), and a
clean-room animation port (phase scroll indicator + a cosine ripple working
line; the spinner was already animated, frozen under ZERO_REDUCED_MOTION).
Reuses the scroll/viewport engine; inline mode unchanged. New unit tests for the
sidebar, plan-clock freeze, and the pure animation helpers; go build/vet/test green.
* feat(swarm): live AGENTS sidebar + un-defer coordination tools
Sidebar AGENTS section now reflects the swarm/team accurately:
- name each member by a short 1-2 word task (not subagent-N)
- a mild, slow cool pulse on live member names (no per-letter flicker)
- members disappear when swarm_status reports them [done]/[failed]
- hide bogus "specialist not found" tool-misroute attempts
- keep the animation tick alive while sidebar agents exist
Plus the deeper fix so a coordinating model stops misrouting swarm calls
to the specialist tool ("specialist swarm_send not found"): the swarm
COORDINATION tools (swarm_send/status/inbox/collect) un-defer once a
swarm is active, while spawn/schedule/handoff stay discoverable via
tool_search. A new DeferralEligible() interface keeps all swarm tools
counting toward DeferThreshold even when exposed, so un-deferring can
never deactivate deferral / force-expose MCP tools in any config.
* feat(tui): polish pass — hierarchy, status line, motion, sidebar toggle
A grounded polish pass across the TUI (17 items):
- color: spread the gray tiers (AA-safe), lime working-ripple (no semantic
green/amber), calm steady ask-label, dim completed plan-step bodies,
bold-muted sidebar headers + stateful PLAN count
- status line: run-state chip (permission mode + effort) replaces the
duplicated provider; composer rule is model-only; context-fill gauge down to
the narrow tier + a sidebar context-% chip
- footer: persistent idle key-hint (managed mode) + a jump-to-bottom cue
- motion: pulsing streaming caret; running specialist spins; swarm/specialist
agents linger with a fading ✓ before removal (no abrupt pop)
- layout: Ctrl+B toggles the sidebar; padded ' │ ' divider; sidebar gated to
the medium tier so it never starves the chat
- home screen: cwd · branch · model orientation + example prompts
- cleanup: drop dead autoTag, fix 'six swarm tools' miscount
go build/vet/test green; -race clean; no hex outside theme.go.
* feat(tui): reading column for the chat transcript
Cap transcript body rows at a readable ~90-col measure and indent them by a
small left gutter, so on a wide terminal the chat text no longer runs
edge-to-edge. Only true transcript rows use the reading column; the frame,
composer, status line, sidebar, title bar, empty state, and prompts keep the
full chatColumnWidth. Streaming text shares the column so it doesn't snap on
finalize. Plus extra vertical spacing before headings and between paragraphs.
Implementation keeps the height cache valid (rows render at contentWidth, which
the width-keyed cache already captures) and the gutter is a horizontal post-pad;
each selectable line's textStart is shifted by the gutter so click-to-select and
the selection highlight stay aligned. Mouse card/toggle hit-testing is Y-based,
so it's unaffected. go build/vet/test green; -race clean.
* fix(tui): plain cancellation marker + copyable card/toggle rows
- 'Run cancelled.' renders as a plain amber '⊘ Run cancelled.' line instead of a
heavy box (other system notices keep their box).
- Copy root cause: the app's text-selection skipped 'button' rows (collapsible
toggles, specialist cards) — they carried no selectable text, so a drag-select
omitted exactly that (dull) content. Fix: those rows now carry their plain text
+ start, so a selection dragged THROUGH them copies their content; a direct
click still toggles/drills (resolved on press, before selection). The selection
highlight no longer excludes them. Permission buttons stay excluded.
- Surface Ctrl+E in the idle hint — it releases the mouse for native terminal
selection, which copies everything (the zero-risk full-copy path).
go build/vet/test green; -race clean.
* fix(tui): render all system notices as plain lines, not boxes
Every system notice (mouse-mode changes, mode set, run cancelled, etc.) now
renders as a plain marked line instead of a bordered box: a faint '·' marker +
muted text for info, amber '⊘' for a run cancellation. Multi-line notices keep
the marker on the first line and indent the rest. Command cards and error rows
keep their boxes. go build/vet/test green.
* fix(tui): separate think→act tool groups with a blank line
Reasoning headers (Thought for X) interleave the tool cards (tool → thought →
tool …), which broke the existing tool-card→tool-card gap rule, so consecutive
list_directory/read_file/update_plan groups packed into a dense wall. Now a
reasoning header that follows a tool card gets a blank line above it, so each
think→act group reads as a distinct block. go build/vet/test green.
* feat(tui): colour markdown emphasis + headings for readability
Assistant prose now reads as three tiers instead of one flat ink colour:
- body text stays ink
- bold / important words render in the brand accent (lime) so they pop
- markdown headings render accent + bold + underline as a clear top tier
go build/vet/test green.
* fix(tui): calmer prose colours + a working-phase label
- Revert the accent colour on bold/important words: emphasis is weight-only now
(no colour), dark-mode-friendly.
- Headings drop the bright lime — distinguished by ink + bold + underline instead
of a saturated accent colour.
- Working status line gains a phase label ('writing' while the answer streams,
'thinking' otherwise) so a long, output-less step reads as live progress
('Working · thinking · 55s') rather than a frozen screen.
go build/vet/test green.
* fix(tui): paint selection highlight in the gutter-shifted coordinate
Mouse text-selection looked like it started in the wrong place: the highlight
landed a few cells right of where you clicked, so selecting/copying felt
impossible to aim.
Cause: the row renderers painted the selection highlight on the UNshifted,
unpadded line, then the body item applied the reading-column gutter (pad lines +
shift selectable textStart) afterward. But the stored selection points are in the
shifted/displayed coordinate (that's how the mouse maps them), so the highlight
was computed against unshifted textStart and landed gutter-cells off. (The copied
text was actually correct — extraction uses the shifted selectable — but the
visual selection was wrong, which is what made it un-aimable.)
Fix: stop highlighting inside the row renderers; add finalizeTranscriptBodyRow,
which pads + shifts THEN paints the highlight, so it's computed in the same
coordinate the mouse uses and lands exactly on the click. Also applies the
highlight uniformly to all rows + the interim block. Regression test added.
|
||
|
KRATOSandgnanam1990
|
f7ee189ba5 |
fix: resolve 15 findings from the 2026-06-20 deep audit (#281)
* docs(audit): 2026-06-20 deep adversarial audit report
Audit-only deliverable (no source changed). Fresh origin/main (
|
||
|
Vasanth T
|
b2bb6c4b41 |
feat: five capability features from the reference-agent comparison (#276)
* feat(commands): user-defined slash commands from markdown files
Drop a markdown file at `.zero/commands/<name>.md` (project, shared with the
team) or `<userConfigDir>/zero/commands/<name>.md` (personal); typing
`/<name> args` expands its body template and submits it as a normal prompt.
Optional YAML frontmatter (`description:`, `model:`, `agent:`) feeds help text
and routing. This turns Zero's model-pulled skills into user-invokable,
repo-checked-in team workflows (`/release v1.2`, `/fix-flaky <test>`).
New internal/usercommands package: Load() scans the project + user dirs
(project shadows user on name collision, mirroring specialist scope
precedence), validates names to a safe slash-command shape, and parses
frontmatter with a self-contained parser. Expand() substitutes $ARGUMENTS,
$1..$9, and $$ (literal $); a placeholder-free template gets the raw args
appended so a trivial command still receives input.
Wiring: model loads commands at construction; the commandUnknown dispatch
checks the user set before reporting "unknown"; autocomplete lists user
commands alongside builtins. Builtins win on collision (listed first, shadow at
dispatch). Tests cover load/precedence/name-validation, the placeholder
substitution table, and the TUI dispatch + autocomplete + still-unknown paths.
* feat(init): guided /init that generates AGENTS.md seeded with repo facts
`/init` (TUI) and `zero init` (headless) investigate the repo and write a
concise AGENTS.md so future runs start with project context. Zero already has
the AGENTS.md loader and the repoinfo scanner, but the file only existed if a
user hand-authored it — this activates the machinery in one command.
New internal/agentinit.BuildPrompt seeds the bootstrap prompt with
repoinfo.Collect() output (primary language, build/test tools, CI, workspace
layout, git) formatted as a fact block, then instructs the agent to investigate
the gaps and write a tight AGENTS.md (build/test commands, conventions, things
to avoid). Collection failure is non-fatal — the agent investigates from
scratch. This is one better than a pure-prompt /init: the agent starts from
known facts instead of re-deriving them.
TUI: /init command + handleInitCommand reuse launchPrompt (standard run, tools,
AGENTS.md write path). Headless: `zero init` builds the prompt and forwards it
through the existing exec machinery, so all exec flags (--model, -w, --add-dir)
work unchanged. Tests cover FormatFacts (rendered/omitted fields), BuildPrompt,
and the TUI dispatch.
* feat(tools): lsp_navigate — semantic code navigation via the language server
A read-only, model-callable tool exposing LSP navigation: jump-to-definition,
find-all-references, find-implementations, and workspace_symbol search. The
agent can answer "where is X defined / who calls X / what implements this
interface" precisely, where grep collides on common names or misses indirect
references — cutting wasted read/grep turns and wrong-edit risk on large repos.
Zero already runs language servers for diagnostics, so this adds the
request/response navigation methods on top: Manager.Navigate syncs the file
(didOpen) then issues textDocument/definition|references|implementation or
workspace/symbol via the existing client.Call, decoding all three result shapes
(Location, []Location, []LocationLink). 1-based file:line:col in and out
(converting LSP's 0-based positions); the tool has its own lazily-started
manager (servers start on first use, reused across calls).
Opportunistic: a file type with no installed server, or a server lacking the
capability, degrades to a clear "unavailable — fall back to grep" message, not
an error. Registered in CoreReadOnlyToolsScoped. Tests cover the response-shape
decoding (incl. LocationLink + null), symbol decoding/kinds, unknown-op and
unsupported-extension degradation, and the tool's arg validation + read-only
safety.
* feat(agent): prune stale tool output before paid summarization
Long, dump-heavy sessions accumulate large read_file/grep/glob/bash tool
results the model has already acted on; their bodies are dead weight the loop
otherwise keeps paying an LLM summarizer to compress. Add a zero-cost prune
stage inside maybeCompact, before the summarizer: when over threshold, replace
the bodies of OLD, large tool results with a compact placeholder, recompute
size, and only fall through to the paid Compact() if still over.
pruneStaleToolOutput walks newest-first, protects the last preserveLast
messages plus a trailing 40k-token window of tool output (most likely still in
use), and prunes only tool results older than that and larger than 200 tokens —
gated so the whole pass is a no-op unless reclaimable output exceeds 20k tokens.
It keeps the tool message and its ToolCallID (provider replay stays valid; the
model knows what was there and can re-run the tool), never touches non-tool
messages, never re-prunes a placeholder (idempotent), and copies-on-write so it
never mutates the caller's slice. Recent turns are preserved verbatim — no
paraphrase loss, no token or latency cost.
Tests cover the reclaim gate, recent-window protection, idempotence,
non-tool-message safety, ToolCallID/role preservation, and no-input-mutation.
* feat(agent): reconnect on a transient model disconnect instead of ending the run
A long autonomous task (a big refactor, a swarm member, a headless/cron run)
should survive a single transient upstream hiccup rather than dying and
re-burning every token on a restart. When the initial StreamCompletion connect
fails with a disconnect-shaped error, streamWithReconnect re-issues the SAME
request with exponential backoff up to 2 times.
It retries only the connect — before any content is forwarded — so no
already-streamed OnText is ever duplicated. A context-limit error (the
compactor's job), an image rejection, a non-disconnect error, or a cancelled
context is returned immediately without retry, so the reconnect never fights
the existing handlers. Disconnect classification matches EOF / connection
reset|refused|closed / broken pipe / timeout / 502 / 503 / server-closed.
Reconnect attempts surface through OnReasoning ("connection lost —
reconnecting N/max"), a non-content channel that never corrupts the answer
text; silent when there's no reasoning sink.
Tests: recover-from-transient, give-up-after-max, no-retry-on-context-limit,
stop-on-context-cancel, the disconnect-vs-not classifier, backoff growth, and
the notice routing.
* fix(specialist): add lsp_navigate to knownToolNames
The new lsp_navigate tool was added to the core read-only registry but not to
specialist.knownToolNames, which mirrors the core tool set — so
TestKnownToolNamesMatchCoreRegistry failed (the list drifted from the registry).
Add it to the read-only group to match.
* fix(tools): confine lsp_navigate to the workspace (close arbitrary-file-read)
Review caught that lsp_navigate dropped the PathScope every sibling read-only
tool (read_file/glob/grep) carries, and its readFile joined or used the
model-supplied path verbatim — so `path:"/etc/passwd"` (absolute) or
`path:"../../etc/..."` (escaping) read arbitrary files off disk AND opened them
into the language-server subprocess, reachable via indirect prompt injection.
Add NewScopedLSPNavigateTool(workspaceRoot, scope) (NewLSPNavigateTool now
delegates with nil scope), resolve the path through resolveScopedReadPath BEFORE
reading or building the LSP request — the same confinement the neighbouring
tools enforce — and hand the LSP manager the resolved absolute path so it never
opens an out-of-workspace file. Errors echo only the relative path, never an
absolute one. Registered via the scoped constructor in CoreReadOnlyToolsScoped.
Regression test asserts `/etc/passwd` and `../../` are rejected for both the
position and workspace_symbol ops; verified by mutation (bypassing the scope
resolution makes the test fail).
|
||
|
Vasanth TandClaude Opus 4.8
|
c572791edf |
feat: clipboard image paste + provider 400 handler for image rejection (#268)
* fix(agent): abort on image-rejection 400s instead of retrying in a loop
* feat(imageinput): add ReadClipboardImage for OS clipboard image extraction
* feat(tui): clipboard image paste — detect and attach screenshots from OS clipboard
* fix(imageinput): use temp file for Windows clipboard image (PowerShell stdout can't emit raw binary)
* test: handle both image-present and image-absent clipboard states
* fix(vision): add kimi-k2 and qwen-vl to vision-capable name heuristic
* fix(tui): render image errors as red error rows instead of grey system notes
* feat(vision): check models.dev InputModalities for vision support instead of only name heuristics
The vision gate now checks three sources:
1. Curated model registry (catalog authority + name heuristic)
2. Discovered model list from models.dev (InputModalities contains 'image')
3. Name heuristic fallback for unknown models
This means custom/ollama/cloud models (kimi-k2, qwen-vl, etc.) are correctly
detected as vision-capable when models.dev reports them with image input
modality — no more manual name matching needed.
* chore: gofmt clipboard_test.go
* fix(imageinput): remove shell injection in the Linux clipboard reader
The Linux clipboard reader built shell commands by concatenating a
clipboard-derived MIME type into `sh -c`:
exec.Command("sh", "-c", "wl-paste --type "+t+" 2>/dev/null")
exec.Command("sh", "-c", "xclip ... -t "+t+" -o 2>/dev/null")
`t` comes from `wl-paste --list-types` / `xclip TARGETS` and is only filtered by
`HasPrefix(t, "image/")`, so a hostile clipboard offerer could register a target
like `image/png; rm -rf ~` — it passes the prefix check and then executes
through the shell. A real (if low-likelihood) local code-exec vector.
Drop the shell entirely: every helper now runs via exec.Command(prog, args...)
with the MIME type passed as a discrete argument (matching the existing no-shell
pattern in pdf.go). Stderr is discarded by not assigning it. Factored into
runClipboardStdout + imageMIMETypes helpers.
Test: imageMIMETypes filters to image/* lines and returns even an
injection-shaped type verbatim (safe now that it never reaches a shell).
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
* test(tools): de-flake TestWriteStdinInterruptTerminatesSession
The test had two Windows-CI flakes that reddened unrelated PRs:
1. It relied on write_stdin's 1000ms yield window being long enough for the
async SIGKILL + reap to land before asserting the session was no longer
running. On a loaded runner the reap exceeded that window, so write_stdin
reported "still running" and the assertion failed. Fix: after terminating,
wait on session.done deterministically (30s safety timeout that fails loudly
if the kill genuinely hangs); the common case returns the instant the
process exits.
2. The SIGKILL'd child had the test's t.TempDir() as its cwd, and Windows may
not release the directory handle the instant the process is reaped — so
t.TempDir()'s immediate RemoveAll on cleanup failed with "being used by
another process". Fix: resilientTempDir retries RemoveAll for up to 5s
(best-effort; a leaked temp dir never fails the test).
Verified 20x locally; full internal/tools suite green.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <[email protected]>
|
||
|
Anandan
|
cf54e2cb0e |
Add persistent exec command sessions (#267)
* Add persistent exec command sessions Introduce exec_command and write_stdin tools backed by a shared session manager so long-running shell commands can remain alive and be polled or interrupted across tool calls. Wire exec_command through the existing shell permission, network prompt, command-prefix approval, sandbox grant scope, specialist manifest, and prompt/documentation paths. Keep web_fetch limited to public remote HTTP(S) URLs so localhost/private URLs go through shell commands where sandbox network permission applies. Tested: git diff --cached --check * Fix exec session lifecycle review issues |
||
|
KRATOS
|
d3e7b89eb0 |
TUI: live code-streaming preview, theme system, plan/Task transcript dedup (#262)
* tui: live code-streaming preview, theme system, plan/Task dedup - Live "writing <file>" preview: a long write_file/edit_file/apply_patch now streams the code as the model generates it (path + line count + a color-coded tail) instead of a frozen spinner. Tool-call argument deltas are forwarded zeroruntime -> agent -> TUI (OnToolCallStart/Delta). - Theme system: auto dark/light detection via a terminal-background probe, a /theme [auto|dark|light] command, a refreshed palette, and streaming fade auto-disabled on low-color / SSH / tmux terminals. - Plan/Task transcript dedup: update_plan collapses to a one-line summary (the live plan panel owns the detail), and Task delegations render only as the specialist card, not also as raw tool-call/result rows -- in both the live and resume paths. Robustness for the streaming preview: tolerate whitespace in tool-call JSON, extract edit_file's new_string, clear the preview on tool finalize / cancel / run-end (no stale or duplicate previews; a stale run can't wipe the active one), color new content green (never red for a leading '-'), and truncate by display width. * address M10: keep failed Task delegations on resume The unconditional Task tool-row skip on resume also hid a Task that failed before a specialist started (there is no card to replace it). A pre-pass now collects the tool-call ids that actually started a specialist; the Task tool-call/result rows are skipped only for those, so a failed delegation and its error stay visible on resume. * streaming preview: decode tool-call args incrementally (O(n) not O(n^2)) The live "writing" preview re-scanned the whole accumulated args buffer on every delta and grew it via string concat, so a large write_file/edit was quadratic and janked exactly during the long write it targets (M14/L6/L7). Replace the per-render decode with a streamingDecoder that consumes each fragment once: it buffers the head until the content value begins (extracting the path), then decodes the content's JSON-string escapes as they arrive, keeping only a bounded tail of lines plus a running line count. The view now reads that pre-computed state in O(tail) per render. Handles whitespace, escapes split across fragments, \uXXXX, and the closing quote; unit-tested including byte-by-byte vs whole-buffer equivalence. |
||
|
Vasanth T
|
b237e9849a |
feat(tui): sticky plan panel + specialist task cards + task table overlay (#255)
* feat(tui): sticky plan panel + specialist task cards + task table overlay Three new TUI surfaces that give users live visibility into agent progress and specialist delegation, so long turns and subagent work never look frozen. Sticky plan panel (plan_panel.go): - Pinned between the title bar and transcript body, shows the agent's step-by-step plan from the update_plan tool as a live checklist. - Header with spinner + progress count + elapsed; text progress bar; per-step list with status icons (✓ completed, ⠹ in_progress, ○ pending, ✗ failed) and individual step timings. - Auto-collapses to a one-line green summary 30s after completion. Ctrl+P toggles expand/collapse. - Plan state synced from update_plan tool results via CurrentPlan(); timestamps preserved across the tool's full-replacement updates by matching on content. Specialist cards (specialist_card.go): - New rowSpecialist transcript row kind; EventSpecialistStart/Stop now handled in session.go hydration (previously fell through to default and were silently dropped on resume). - Animated card with spinner + name + task + elapsed + token count + tool call count. Blue while running, green when done, red on error. - specialistTracker tracks live state; Task tool calls/returns are intercepted in the OnToolCall/OnToolResult callbacks to start/complete tracking and emit card rows. - Height cache marks running specialist rows unstable so the spinner animates. Task table overlay (task_table.go): - Ctrl+G toggles a full-width overlay showing all specialists in the session as a sortable table: #, Specialist, Task, Status, Time, Tokens, Tools. - Summary bar with running/completed/error counts, total tokens, aggregate elapsed. - Arrow keys navigate rows; Esc closes the overlay. Tests: plan_panel_test.go (7 tests), specialist_card_test.go (11 tests), task_table_test.go (8 tests). All existing TUI tests pass. * fix(update_plan): expose structured nested item schema so models stop sending bad arguments The plan property's Items schema was omitted because of an outdated comment claiming the serializer drops nested Properties/Required. The serializer (propertyToRuntimeMap) already handles them recursively. Without the nested schema, weaker models guessed the item structure from a text description and sent malformed arguments, hitting the 4-strike guardrail halt. Now exposes Items as an object with content (string, required), status (enum), and notes (string). * fix(tui): route plan + specialist updates through message sink, not stale closures The OnToolCall/OnToolResult callbacks captured model by value, so m.plan.updateFromItems() and m.specialists.start/complete() mutated stale copies that never reached the live model. Introduced planUpdateMsg, specialistStartMsg, and specialistCompleteMsg tea.Msg types sent through runtimeMessageSink, handled in updateModel() where they mutate the live model. * feat(tui): subchat drill-in for specialist sessions Enter on a task table row loads the child session's events and swaps the transcript view to show the subagent's full conversation (user messages, agent text, tool calls, results). A nav bar at the top shows the specialist name/task and a back hint. ArrowUp or Esc pops back to the main chat at the saved scroll position. - subchat.go: subchatState with enter/exit, renderSubchatNavBar - transcript_selection.go: transcriptBodyItemsFromRows renders arbitrary rows (reuses the same rendering pipeline as the main transcript) - model.go: Enter on task table row triggers drill-in; ArrowUp/Esc exits; transcriptView() short-circuits to child rows when subchat is active * fix(ci): strip CRLF line endings and BOM from plan_tool_test.go and update_plan.go gofmt and git diff --check failed on ubuntu CI because PowerShell's Set-Content wrote UTF-8 with BOM and CRLF. Stripped both to plain LF without BOM. * feat(tui): add renderLeftRuleCard helper for borderless specialist cards * feat(tui): switch specialist cards to left-rule rendering, drop hint line * feat(tui): add renderSpecialistSummary one-line rollup for specialist cards * fix(tui): correct slice offset in renderSpecialistSummary and add pluralization test The muted tail was sliced at summary[len(spinnerView)] but the spinner sits at byte offset 2 (after the 2-space indent), so the slice cut into the spinner's UTF-8 continuation bytes and dropped the indent. Slice at 2+len(spinnerView) instead. Also add a test asserting '2 errors' is present and '2 error' (singular) is absent when two specialists errored. * refactor(tui): remove task table overlay — cards carry everything now * feat(tui): render specialist summary line above cards in transcript * feat(tui): mark specialist card lines as clickable selectable rows * feat(tui): click specialist card to drill into subchat, Enter as keyboard fallback * tui: clear selectedSpecialistID after Enter drill-in Fixes Task 7 review issue: selectedSpecialistID persisted after Enter drill-in, causing a subsequent Enter to re-drill into the specialist's subchat instead of submitting the composer. Clear it immediately after subchat.enter succeeds so the keyboard fallback is one-shot per click; click re-sets it, so click->Enter still works. * chore: gofmt model.go + test files * chore: gitignore .superpowers/ and docs/superpowers/ — local skill artifacts * fix(tui): remove Enter keyboard fallback for subchat drill-in — mouse click only * fix(tui): reconcile specialist childSessionID from tool call ID to real session ID on completion * feat(specialist): add Progress callback field to RunOptions and TaskRunOptions * feat(specialist): stream child stdout line-by-line via Progress callback * feat(agent): wire OnToolProgress callback for specialist child events * feat(tui): show live tool-call progress in specialist cards * fix(swarm): update RunChild mock signature for new progress callback param * fix(tui): render plan panel inline in transcript, not pinned at top * chore: gofmt model.go, specialist_card.go, specialist_click_test.go * fix(tui): address PR review - plan panel without specialists, stream reader safety, resume dedup + payload keys |
||
|
Anandan
|
ac6addaa83 |
Align sandbox permission blocks and backend states (#253)
* Prompt for shell network permissions Route network-risk shell calls through the existing permission prompt instead of hard-denying them at sandbox preflight. Approved network prompts apply the current turn/session request-permissions overlay so the effective sandbox policy is widened only after user approval. Move network command detection into the shell AST analyzer for package installers, gh/git network subcommands, and python http.server so quoted process patterns such as pkill -f "python3 -m http.server 8000" are not misclassified as network activity. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox -run 'TestAnalyzeCommand|TestClassifyDoesNotFlagQuotedHttpServerPattern|TestEvaluateExemptsNetworkToolsFromDenyByDefault|TestEngineClassifiesNetworkAndDestructiveShellCommands|TestGrantRequestPermissionsNetworkOverlaysPolicyForTurn' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent -run 'TestRunPersistentCommandPrefixStillPromptsForNetwork|TestRunApprovedNetworkBashPromptAppliesTurnNetworkGrant' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/tools -run 'TestBash|TestRegistryRunsBashThroughSandboxEngine' Tested: GOCACHE=/tmp/zero-go-cache go test ./... * Auto-allow native-sandboxed shell commands Remove the old hidden bash auto-allow environment/config path. Ordinary shell commands now use the active native sandbox as the permission boundary by default, while network, destructive, and workspace policy checks still run first. Policy-only and disabled sandbox backends continue to require approval for shell commands, so unsandboxed execution is not silently allowed. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox -run 'TestSandboxedBash|TestShellSandboxActive|TestEvaluateExemptsNetworkToolsFromDenyByDefault|TestGrantRequestPermissionsNetworkOverlaysPolicyForTurn|TestAnalyzeCommand|TestClassifyDoesNotFlagQuotedHttpServerPattern' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/tools -run 'TestBashAuto|TestBashStillPromptsWithoutActiveSandbox|TestRegistryRunsBashThroughSandboxEngine' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/cli -run 'TestApplyConfiguredSandboxPolicy|TestApplyConfiguredAutonomyCeiling' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent -run 'TestRunPersistentCommandPrefixStillPromptsForNetwork|TestRunApprovedNetworkBashPromptAppliesTurnNetworkGrant' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/daemon -run 'TestScrubWorkerEnvRemovesReentrancyMarkers' Tested: GOCACHE=/tmp/zero-go-cache go test ./... * Prompt for destructive shell approval Remove the separate DenyDestructiveShell policy layer from active sandbox decisions. Destructive shell classification now produces an approval prompt in ask/auto flows instead of a sandbox hard-deny, while no-approval paths remain blocked. Drop the removed destructive-shell policy field from sandbox policy output, backend capabilities, snapshots, and Windows golden data. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent -run 'TestRunPromptsForDestructiveShellInsteadOfSandboxDeny|TestRunPersistentCommandPrefixStillPromptsForNetwork|TestRunApprovedNetworkBashPromptAppliesTurnNetworkGrant' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox -run 'TestEngineClassifiesNetworkAndDestructiveShellCommands|TestBackendCapabilitiesReflectDisabledPolicy' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/cli -run 'TestRunSandboxPolicy|TestEffectiveSandboxPolicy' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/zerocommands -run 'TestSandboxPolicySnapshotFromPolicyFillsEffectiveMode' Tested: GOCACHE=/tmp/zero-go-cache go test ./... * Remove web-tool network policy switch Remove the old EnforceToolNetwork policy option and the NetworkHostAllowed per-tool gate so sandbox network policy only controls sandboxed shell egress. Keep web_search and web_fetch on their own tool permission and URL safety paths, with tests proving deny/scoped shell network policy no longer blocks in-process web tools. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox -run 'TestEvaluateExemptsNetworkToolsFromShellNetworkPolicy|TestEngineAllowsNetworkSideEffectWhenShellPolicyBlocksNetwork|TestGrantRequestPermissionsNetworkOverlaysPolicyForTurn|TestScopedNetworkGateInEvaluate' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/tools -run 'TestWebSearchRunWithSandbox|TestWebFetchRunWithSandbox|TestRegistryRoutesWebFetchThroughSandboxPath' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent -run 'TestRunPersistentCommandPrefixStillPromptsForNetwork|TestRunApprovedNetworkBashPromptAppliesTurnNetworkGrant' Tested: GOCACHE=/tmp/zero-go-cache go test ./... Tested: git diff --check * Prompt for outside-workspace file access Convert recoverable outside-workspace file path violations into permission prompts instead of hard-denying in ask/auto modes. Approved prompts now grant the requested filesystem scope through the existing request_permissions overlay for the current turn or session, so the follow-up sandbox evaluation and scoped file tools use the same policy path. Keep unsafe mode, apply_patch escapes, symlink traversal, and non-recoverable path cases hard-denied. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox -run 'TestEngineAutoAllowsWorkspaceFileMutationTools|TestEngineDeniesOutOfWorkspacePaths|TestGrantRequestPermissionsReadDoesNotGrantWrite|TestGrantRequestPermissionsSessionPersists' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent -run 'TestRunPromptsAndAllowsOutsideWorkspaceWrite|TestRunSessionAllowsLaterOutsideWorkspaceWrite|TestRunAppliesSandboxEvenInUnsafeMode' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/tools -run 'TestRegistryAppliesSandboxBeforeToolExecution|TestRegistrySandboxGatesPathAliasKeys|TestRegistryAllowsPromptToolWithPersistentSandboxGrant' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/cli -run 'TestRunSandboxCheckJSONDeniesOutOfWorkspaceWrite' Tested: GOCACHE=/tmp/zero-go-cache go test ./... Tested: git diff --check * Fix Windows network prompt test fixture Use a Windows curl.cmd shim and cmd.exe PATH syntax for TestRunApprovedNetworkBashPromptAppliesTurnNetworkGrant while keeping the POSIX fake curl fixture for Linux and macOS. This keeps the test focused on network prompt approval and turn network grants instead of relying on a POSIX shell script being executable on Windows. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent -run 'TestRunApprovedNetworkBashPromptAppliesTurnNetworkGrant|TestRunPersistentCommandPrefixStillPromptsForNetwork' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox -run 'TestClassifyASTCatchesNetworkProgramsRegexMisses|TestClassifyFlagsPipedInstallerVariants' Tested: GOCACHE=/tmp/zero-go-cache go test ./... Tested: git diff --check * Remove sandbox autonomy ceilings Sandbox permission decisions now match grants by tool and scope only, without MaxAutonomy or request autonomy checks. Persistent/session grants, permission approvals, workspace-write auto-allow, and native-sandbox shell auto-allow no longer route through an autonomy ceiling. Remove the sandbox.maxAutonomy config and CLI surfaces, including grant --auto, policy output, sandbox check autonomy input, and sandbox grant snapshots. MCP permission autonomy is intentionally unchanged because it is a separate MCP permission model. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox ./internal/agent ./internal/tools ./internal/zerocommands ./internal/config ./internal/cli ./internal/tui Tested: GOCACHE=/tmp/zero-go-cache go test ./... * Fail closed without native sandbox Remove AllowPolicyOnlyRunner from sandbox policy and snapshots. A restricted/managed command plan now errors when native sandbox support is unavailable instead of running through a policy-only direct runner. Policy-only backend data remains as diagnostics for capability/reporting paths, but it no longer authorizes command execution when a platform sandbox is required. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox ./internal/zerocommands ./internal/cli ./internal/tui Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent ./internal/tools Tested: GOCACHE=/tmp/zero-go-cache go test ./... * Remove scoped shell network policy Drop the domain-scoped shell network mode and the local scoped-egress proxy path from the active sandbox model. Shell network policy now serializes and evaluates only deny or allow. Request-permissions network grants widen to allow without domain lists, and platform profiles consume the binary mode directly. The Linux helper no longer accepts proxy-route flags, Windows network hashes only the binary mode, and web tool tests assert they remain separate from shell network policy. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox ./internal/tools ./internal/notify Tested: GOCACHE=/tmp/zero-go-cache go test ./... * Align permission grant ordering and prompt text Move persistent and session grant matching ahead of recoverable sandbox prompts while keeping explicit denies and non-recoverable path violations fail-closed. Clean TUI permission transcript/card text so normal prompts show action, reason, scope, and readable block details instead of raw permission/risk/violation key-value metadata. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox ./internal/tui ./internal/agent ./internal/tools Tested: GOCACHE=/tmp/zero-go-cache go test ./... * Remove policy-only sandbox terminology Replace the unavailable native-backend diagnostic path with explicit unavailable support/backend states, and stop presenting missing native isolation as a policy-only fallback. Rename structured sandbox denial metadata from violation to block across permission events, stream-json, snapshots, TUI rendering, and sandbox-check output so user-facing flow matches blocked action language. Update README, work split docs, smoke test selectors, and Windows sandbox policy golden output for the new backend/support names and block payloads. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox ./internal/agent ./internal/tools ./internal/tui ./internal/cli ./internal/doctor ./internal/zerocommands ./internal/streamjson Tested: GOCACHE=/tmp/zero-go-cache go test ./... Tested: scripts/sandbox-smoke.sh Tested: git diff --check * Fix sandbox review follow-ups Fail closed for unknown network modes, keep network prompts ahead of bash allow grants, preserve v1 grant-file reads, and classify unparseable commands with obvious network programs as network-risk. Bump stream-json schema to v2 and document the block metadata field after the violation-to-block rename. Tests: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox ./internal/streamjson ./internal/specialist ./internal/cli -run 'TestExec.*Stream|TestStreamJSON|TestExecProtocol|TestExecReadsStreamJSONImages|TestExecAcceptsImageOnlyStreamJSONMessage|TestExecInputFormatStreamJSON' Tests: GOCACHE=/tmp/zero-go-cache go test ./... Tests: scripts/sandbox-smoke.sh Tests: git diff --check |
||
|
Anandan
|
673fce11bd |
Persist command prefix approvals (#252)
Add file-backed command prefix grants to the sandbox store and load them through the engine so approved prefixes skip future prompts while still running through sandbox policy checks. Tighten reusable prefix validation for runner-style launchers and script paths, and classify find -delete as destructive. Surface persistent prefix choices in the TUI and include approved prefixes in the run prompt context and sandbox grants list. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent ./internal/sandbox ./internal/tui ./internal/cli Tested: GOCACHE=/tmp/zero-go-cache go test ./... Tested: git diff --check |
||
|
Anandan
|
f00e83a5ed |
Add session command prefix approvals (#250)
Bash permission prompts can now expose a validated session-scoped prefix approval. Prefixes are parsed from simple shell commands or an explicit prefix_rule argument, with redirects, control operators, assignments, expansions, globs, and broad launcher prefixes rejected before the option is shown. Approved prefixes are stored in the session sandbox engine and only skip future matching bash prompts; commands still execute through the existing sandbox backend. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent ./internal/tui Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent ./internal/sandbox ./internal/tui ./internal/tools Tested: GOCACHE=/tmp/zero-go-cache go test ./... |
||
|
Anandan
|
2cf6fd2334 |
Add request permissions flow (#248)
* Add request permissions flow Adds a runtime-handled request_permissions tool that can ask for turn or session filesystem/network grants and return the granted profile as structured JSON. Temporary turn grants update the shared sandbox scope and effective policy, then clean up at run end; session grants persist in memory for the session. Read grants now stay read-only through tool path resolution, policy evaluation, and native sandbox profile construction. The TUI renders request_permissions with grant-specific choices, including strict review, and Esc continues without granting permissions. Tested: GOCACHE=/tmp/zero-go-cache go test ./... * Align request permission grants with prompts Display and return the normalized grant roots that the sandbox actually applies, so file-like permission requests no longer show a narrower path than the directory root being granted. Also makes the network prompt explicit about all outbound access and relabels the strict option as a model review request instead of implying enforced command review. Tested: GOCACHE=/tmp/zero-go-cache go test ./... * Retry transient OAuth secret read locks Windows can briefly deny reads of the freshly published token secret while concurrent OAuth secret creation converges. Treat Windows sharing and lock violations as transient and retry those reads before returning an error. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/oauth Tested: GOCACHE=/tmp/zero-go-cache GOOS=windows GOARCH=amd64 go test -c ./internal/oauth -o /tmp/zero-oauth-windows.test.exe Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox ./internal/agent ./internal/tui ./internal/tools ./internal/specialist ./internal/oauth Tested: GOCACHE=/tmp/zero-go-cache go test ./... |
||
|
KRATOS
|
995b9086c7 |
feat(subagent): proactive specialist delegation + swarm execution fixes (#243)
* feat(subagent): proactive specialist delegation + swarm execution fixes Fixes: - Task tool dropped the parent's permission mode, so every delegated sub-agent ran at --auto high regardless of the parent. Forward PermissionMode so a child never exceeds its parent's authority. - Swarm members never executed: swarm_spawn accepts agent types teammate/subagent, but the launcher looked those up as specialist names (only worker/explorer/code-review exist), so every member failed with "specialist not found". Run each member from its swarm definition via an inline manifest on the executor. Enhancements: - Auto-delegation: the orchestrator system prompt now lists the available specialists and when to use each, nudging it to offload read-heavy work to a sub-agent so large tool output stays out of the main context. - @-mention: a leading "@<specialist> <task>" in the TUI composer delegates to that specialist (transcript keeps the verbatim mention); the @ palette suggests specialists on the first token, file references mid-message. - swarm.maxTeamSize config caps concurrent members per team (0 => default 8; negative rejected), honored in user and project config. - Proactive spawning: read-only specialists (explorer/code-review) auto- approve via a new args-aware tool permission (tools.ArgsPermissioner), so the agent delegates exploration off a normal prompt without an approval prompt; write-capable specialists, resume, and unknown names still prompt. The exec gate registers specialist tools at medium/high autonomy so a non-trivial headless run can auto-delegate read-only too. * fix(specialist): fail-safe child autonomy when PermissionMode is unset specialistAutonomy mapped an empty PermissionMode to "--auto high" (unsafe), silently re-introducing the privilege escalation the rest of this PR fixes: an orchestrator that forgets to wire PermissionMode would get an unsafe child. The Task tool and swarm both now propagate a resolved mode, so "" only occurs for a caller that omitted it -- default that case to "low". Only an explicit unsafe parent now yields --auto high; everything else (incl. unset/unknown) is fail-safe low. Updates the stale "back-compat" comments and the tests that documented the old behavior. Addresses the non-blocking follow-up from PR review. |
||
|
Anandan
|
eb608741b0 |
Align permission prompt rendering (#246)
* Align permission prompt rendering Remove normal permission prompt risk labels from focused cards, transcript rows, and permission detail text so risk metadata is not presented as the permission UI contract. Rename the recoverable deny option from a redirect-style prompt to a continue-without-running label that matches its current behavior. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/tui Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/tui ./internal/agent ./internal/cli Tested: GOCACHE=/tmp/zero-go-cache go test ./... Tested: git diff --check * Tighten permission prompt fallbacks Use conservative allow/deny choices when a permission request lacks backend-supplied decisions, avoiding impossible session or persistent grant options. Document that a failed session-grant recording attempt must not deny the current user-approved tool call; it only means a later matching call prompts again. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/tui ./internal/agent Tested: GOCACHE=/tmp/zero-go-cache go test ./... Tested: git diff --check * Split permission deny and cancel decisions Add an explicit cancel permission decision that aborts the active run while keeping deny as a recoverable tool refusal. Expose cancel choices for command and apply-patch approval prompts, keep apply-patch off the recoverable deny path, and stop offering persistent always approvals for command/apply-patch prompts until scoped prefix approval exists. Classify cancel decisions as permission-decision events for TUI session logs and exec JSON output. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent ./internal/tui ./internal/cli Tested: GOCACHE=/tmp/zero-go-cache go test ./... Tested: git diff --check * Stabilize OAuth secret creation Publish encrypted OAuth token secrets atomically under a short-lived creation lock so concurrent first-run callers never observe a half-written secret file. This fixes the Zero Review failure in TestLoadOrCreateSecretConcurrentConverges. Tested: go test ./internal/oauth -run TestLoadOrCreateSecretConcurrentConverges -count=100 Tested: go test ./internal/oauth Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/agent ./internal/tui ./internal/cli Tested: GOCACHE=/tmp/zero-go-cache go test ./... Tested: git diff --check |
||
|
Anandan
|
2d62387e08 |
Align permission session approvals (#244)
* Align permission session approvals Add in-memory session grants, workspace-write auto-allow for built-in file mutation tools, and dynamic permission choices for once/session/future approvals. Persist future approvals only when a grant store is configured, while keeping session approvals scoped to the active engine. Tested: GOCACHE=/tmp/zero-go-cache go test ./... * Guard workspace write auto-allow Exclude protected workspace metadata from built-in file-tool auto-allow so writes under .git, .zero, and .agents re-prompt instead of silently running. Derive apply_patch targets from patch headers inside the sandbox engine so patch-body escapes and symlink traversal are denied before workspace auto-allow. Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox -run 'TestEngineAutoAllowsWorkspaceFileMutationTools|TestEngineDoesNotAutoAllowProtectedMetadataWrites|TestEngineDeniesApplyPatchEscapesFromPatchBody|TestEngineSessionGrantDoesNotMatchSiblingFile|TestEngineDeniesAutoAllowedSymlinkEscape|TestEngineDeniesWorkspaceSymlinkTraversal|TestDeriveScope' Tested: GOCACHE=/tmp/zero-go-cache go test ./internal/sandbox ./internal/agent ./internal/tui ./internal/tools Tested: GOCACHE=/tmp/zero-go-cache go test ./... Tested: git diff --check |
||
|
Anandan
|
a1b9bbb07c |
Align permission prompts for web tools (#242)
Treat hosted web_search as an allowed read-only network discovery tool so normal searches do not stop on a permission prompt, while retaining sandbox network enforcement when tool-network policy is enabled. Add host-scoped persistent grants for web_fetch URLs so an always-allow decision is bounded to the requested host instead of becoming a blanket tool-wide fetch grant. Update permission prompt and transcript rendering to show network targets, scoped always labels, and preserved scope metadata across live rows, render caching, and session hydration. Tested: GOCACHE=/tmp/zero-go-cache go test ./... |
||
|
Anandan
|
d1c5af7392 |
Polish model status and reasoning streams (#190)
* Polish model status and reasoning streams
* Polish custom provider setup flow
* Clean up pending composer hints
* Address custom provider setup review
* Polish composer overflow handling
* Fix composer paste and wheel edge cases
* Use line counts for paste previews
* Revert "Use line counts for paste previews"
This reverts commit
|
||
|
KRATOS
|
474f494b93 |
Deep-audit remediation + sandbox hardening (batches 1–35) (#189)
* audit batch 1: redact nested secrets fully; count image tokens in compaction estimate
- secrets.Redact: replace longest matches first so a containing PRIVATE KEY
block is redacted before a nested shorter match corrupts its exact string
(audit medium internal/secrets/scanner.go:81) + regression test.
- agent compaction estimateTokens: add a flat per-image token cost so an
image-heavy context trends toward compaction instead of reading as ~0
(audit medium internal/agent/compaction.go:58) + test.
- ledger: re-verify against current main; mark scanner.go:35 and
redaction.go:362 already-fixed-on-main, record this batch's fixes.
* audit batch 2: worktrees defaultRunGit separates stdout/stderr
Capture git stdout and stderr into separate buffers instead of CombinedOutput,
so parsed rev-parse output is not polluted by git warnings and CommandResult.Stderr
carries the real error text (audit medium internal/worktrees/worktrees.go:247) + test.
* audit: re-verify all 87 pending findings against current main
Independent agent per file re-checked each open/partial finding against the
actual current code (the 06-11 ledger was verified on a since-diverged branch),
with an adversarial second pass on every already_fixed verdict. Authoritative
result: 78 genuinely pending (66 still_open + 12 partial); 9 rows were already
fixed on main and are reclassified. Full table in 2026-06-13-reverification.md.
* audit batch 3 (agent): cancellation identity, always-allow w/o sandbox, compaction usage, dedup
- loop.go: check ctx.Err() before collected.Error so a cancelled run returns
context.Canceled (errors.Is holds) instead of a plain stringified error.
- loop.go: 'Always allow' with no sandbox engine now allows THIS call instead of
denying it (persistence is skipped when there's nowhere to persist).
- loop.go: remove a duplicate schema["items"] assignment in propertyToRuntimeMap.
- compaction.go: forward OnUsage (token accounting) through the summarizer while
still omitting OnText, so compaction's token cost is counted in usage/budgets.
- tests for all three behavioral fixes (each fails pre-fix).
* audit batch 4 (cli): list-tools -o json, interrupted -o json terminal, surface session-record failures
- exec --list-tools -o json now emits a JSON tool listing instead of falling
through to the human-readable text (audit medium exec.go:182) + test.
- interrupted run with -o json now emits a terminal error+done on stdout instead
of only printing to stderr, so a JSON consumer sees a clean end (medium exec.go:500) + test.
- execSessionRecorder: surface a latched session-append failure to stderr once at
run end (deferred) instead of silently dropping it (medium exec_sessions.go:114/115) + test.
- left app.go trailing-arg ignore as-is: documented-intentional, TUI has no argv prompt.
* audit batch 5 (zerogit): status -z porcelain (renames + non-ASCII paths); document env-drop on plain Runner
- parseStatus: switch `git status` to `--porcelain -z` and parse the
NUL-delimited form (audit medium zerogit.go:212). Previously the default
--short output was split on newlines, so a rename `R old -> new` was kept as
one bogus path and a non-ASCII/whitespace filename arrived C-quoted+escaped
(e.g. `"caf\303\251.txt"`). The -z form never quotes paths and emits a
rename/copy as `XY <dest>\0<src>`, so we now record the destination verbatim
and consume the source field. + TestParseStatusZHandlesRenamesAndSpecialPaths.
- resolveRunners: document that the plain-Runner EnvRunner adapter intentionally
drops env (audit low zerogit.go:417). The branch is reached only when a caller
supplies a custom Runner without a RunGitEnv — in practice fake-Runner tests.
Production leaves both nil and gets defaultRunGit/defaultRunGitEnv, which honor
GIT_INDEX_FILE, so snapshot-diff isolation holds. Routing env calls to real git
instead would break the fake-Runner test pattern and fix nothing in production.
- update Inspect/Commit mock tests to the -z status format and the new command.
* audit batch 6 (config): reject negative maxTurns; drop dead contract-gap helpers
- FileConfig.UnmarshalJSON now rejects a negative maxTurns instead of letting the
`MaxTurns > 0` merge gates silently drop it and fall back to the default,
hiding the misconfiguration (audit medium resolver.go:136). 0 is left as the
"unset" sentinel (omitempty) and still falls back to the default; the CLI
--max-turns flag continues to reject 0 too since there it is distinguishable
from absent. + TestResolveRejectsNegativeMaxTurns / ...ZeroFallsBackToDefault.
- remove dead config/contracts.go (ContractGap, DefaultContractGaps,
FindContractGapsByMilestone, ContractOwnerRuntime) and its test — referenced
only by that test, never by production (audit low contracts.go:12).
- NOT changed: ResolveMCP intentionally does not run the provider command
(TestResolveMCPDoesNotRunProviderCommand guards this), so the "ZERO_PROVIDER_COMMAND
MCP servers dropped" row is by-design, not a defect.
* audit batch 7 (mcp): bound captured MCP server stderr
connectStdio attached a plain bytes.Buffer as cmd.Stderr that is only ever read
on initialize failure, yet it kept growing for the entire process lifetime as a
long-lived server logged to stderr (audit medium client.go:98). Replace it with
a concurrency-safe boundedBuffer capped at 64 KiB that keeps the head (enough for
init-error diagnostics) and discards the rest, reporting full write lengths so
os/exec never sees a short write. + TestBoundedBufferCapsRetainedBytes.
Left as-is (not defects worth the risk in the MCP client):
- client.go writer.write under client.mu: already designed to release the lock
before the unbounded response wait; the residual concern is a server that stops
draining stdin, an involved async rewrite for a theoretical deadlock.
- readLoop not answering server-initiated requests (ping): responding safely
needs re-locking around the shared writer; no configured server relies on it.
* audit batch 8 (tools/providers): read_file truncation marker; drop dead provider branch
- read_file: when a max_lines cut occurs, append an explicit "[truncated: N more
line(s)... set start_line=X to continue]" marker to the output. Previously only
the Result.Truncated flag was set, which is invisible in the rendered text, so
the model could not distinguish a cut read from a complete one (audit low
read_file.go:89). + TestReadFileToolMarksTruncation.
- providers/factory: simplify `if !explicitProvider || isImplicitOpenAI(...)` to
`if !explicitProvider` and delete the now-unused isImplicitOpenAI. The clause
was dead: explicitProvider==true implies ProviderKind or Provider is set, but
isImplicitOpenAI required both empty, so it could never contribute a case
(audit low factory.go:118). Behavior-preserving; providers tests still green.
* audit batch 9 (sessions): Tree builds from a single List() snapshot (no N+1 disk scans)
Store.Tree recursed via ListChildren per node, and each ListChildren ran a full
store.List() that re-read every metadata.json from disk — O(nodes * total-sessions)
reads for one tree (audit medium lineage.go:140). Snapshot all sessions once, index
children by parent in memory, and recurse over that. Child ordering is factored into
sortChildSessions so ListChildren and Tree stay identical; path-scoped cycle
detection (delete(seen,...)) is preserved. Existing tree/lineage tests still pass.
Left as-is: store.go timestamp() second precision (RFC3339) — switching to
RFC3339Nano would break lexical UpdatedAt ordering against existing second-precision
records ("...00.5Z" sorts before "...00Z"), so it is not a safe one-line change.
* audit batch 10 (specialist): serialize accounting dedup-then-append
recordSpecialistStop and appendSpecialistUsageRollup checked for an existing event
(specialistEventExists → ReadEvents) and then appended under SEPARATE store locks,
so two concurrent finishers — a foreground onExit racing a TaskOutput poll, or a
background reaper — could both pass the dedup check and write duplicate stop/usage
events (audit medium accounting.go:72). Guard each check-then-append pair with a
package-level accountingMu (Executor is used by value, so a struct field would not
be shared). + TestRecordSpecialistStopDedupesUnderConcurrency (passes under -race).
* audit batch 11 (zeroruntime/gemini): O(1) text accumulation; gemini cached-token counts
- zeroruntime CollectStreamWithOptions accumulated assistant text with
`collected.Text += event.Content`, reallocating the whole string on every chunk
(O(n^2) over a long streamed response). Accumulate in a strings.Builder on the
collector and materialize it once in flush, the single finalization point hit
before every return (audit low helpers.go:104). Existing collect tests still pass.
- gemini decoded usageMetadata without cachedContentTokenCount, so a cached-prompt
response reported zero cached input tokens while the other providers report them
(audit medium gemini/types.go:88). Decode the field and thread it through
streamState into NormalizeUsage as CachedInputTokens. + extended
TestStreamCompletionEmitsTextUsageAndReasoningTokens.
* audit batch 12 (specialist): parse child stream without a line-length cap; drop dead helper
- ParseStream used a bufio.Scanner capped at 1 MiB, so one large stream-json line
from a child (a big tool result or final answer) hit bufio.ErrTooLong and aborted
the entire specialist run (audit medium streamer.go:43). Switch to a bufio.Reader
with ReadString, which has no per-line limit; the child is our own trusted
subprocess, so its output length is the legitimate bound. Line numbering and
blank-line/EOF handling are preserved. + TestParseStreamHandlesLineLargerThan1MiB.
- remove dead formatTaskOutput (output_tool.go:244): zero callers and zero tests;
it only forwarded to formatTaskOutputSummary, which is the one actually used.
* audit batch 13 (background/specialist): kill background tasks as a process group
Background children were launched in the parent's process group and terminated by
signalling only the single PID, so any process the task forked was orphaned and
outlived a cancel; the signal also risked hitting a recycled PID after the child
was reaped (audit medium process_posix.go:38/50).
- ConfigureChildProcessGroup (Setpgid on POSIX, no-op on Windows where taskkill /T
already tree-kills) is applied at launch so the child leads its own group.
- terminateProcess now SIGTERM/SIGKILLs the negative PID (the whole group), with a
pid>1 guard so a bogus 0/1 can never expand to "our own group" or "everything",
and treats ESRCH as already-gone. Liveness polls the group. Targeting a group
(rarely recycled vs. a bare PID) also shrinks the reaped-then-recycled window.
- specialist launchBackgroundProcess opts every child into its own group; on a
SetPID failure it now kills the child + group, marks the task errored, and records
the stop (deduped) instead of leaking an untracked orphan.
- tests launch as group leaders; new TestTerminateProcessKillsForkedChildren proves
a forked grandchild dies with the group.
* audit batch 14 (sandbox): quote-aware shell segment splitting
splitShellSegments used a quote-unaware strings.Replacer, so a shell operator
inside quotes was treated as a separator: `git commit -m "use top | less"` split
into a `less` segment and `echo "a; vim b"` into a `vim` segment, both falsely
flagged as interactive (audit medium safe_command.go:138).
Replace it with a quote-aware scanner that mirrors the shell: single quotes make
everything literal; double quotes keep $(...)/`...` substitution active but treat
|, ;, && as literal; unquoted text splits on the same operator set as before
(;, |, ||, &&) plus substitution boundaries. Real unquoted operators still split,
so an interactive program behind a separator — or inside a live `"$(...)"`
substitution — is still caught (no new false negatives, the security-relevant
direction). + TestDetectInteractiveQuoteAwareSeparators covering both directions.
* audit batch 15 (cron): collapse DST fall-back repeated hour to one fire
On a fall-back day the clock repeats an hour (e.g. 01:30 EDT then 01:30 EST), so
Next("30 1 * * *", <01:30 EDT>) returned the same-day 01:30 EST — firing a daily
job twice (audit medium schedule.go:171, which previously guarded only the
spring-forward gap).
Next now skips a match whose local wall-clock minute equals `after`'s but whose
absolute time is later (the repeated-hour duplicate) and keeps searching. Normal
forward search always advances the wall-clock, so this only triggers on the
fall-back repeat; when the scheduler passes the last fire time as `after`, the
repeated hour collapses to a single fire. + TestNextDSTFallBackDoesNotRepeatHour
(spring-forward termination test still passes).
Note: Next is stateless, so it dedupes relative to `after`; a caller that passes
an arbitrary mid-hour "now" instead of the last fire time is out of scope here.
* audit batch 16 (mcp): skip SSE notifications before the response; larger SSE buffer
- decodeSSERPCMessage returned the first non-empty `message` event unconditionally,
so a server-initiated notification/request arriving before the response made the
caller see an id mismatch and fail (audit medium network_client.go:621). It now
skips any message carrying a method (notifications/requests have one; a response
does not) and returns the first real response; the caller's rpcIDMatches still
validates the id. + TestDecodeSSERPCMessageSkipsNotifications.
- raise the per-event SSE scanner cap from 1 MiB to a bounded 8 MiB so a large but
legitimate MCP message no longer hits bufio.ErrTooLong and fails the request
(audit medium network_client.go:637); still bounded against an unbounded server.
Out of scope (kept): full reconnect of a persistent SSE stream after a fatal read
error — a larger change than this remediation.
* audit batch 17 (sessions): fsync session metadata before the atomic rename
writeMetadata wrote the temp file with os.WriteFile (no sync) and renamed it into
place, so a crash right after the rename could expose a metadata file whose bytes
were never flushed — a torn or empty file that corrupts the whole session (audit
medium store.go:632). Write+fsync the temp via a new writeFileSync helper before
renaming, closing that window. + TestWriteFileSyncRoundTrips.
Deliberately scoped:
- Event-log appends (appendEventLocked) are NOT fsynced per append: that would add
a disk flush to every event (every tool call), and a hard-crash loss of the log's
unsynced tail is acceptable, whereas torn METADATA is not. Metadata is the
integrity-critical part and is what this syncs.
- Directory fsync is omitted for cross-platform simplicity (it errors on Windows);
the file fsync already prevents the torn-contents failure, the serious case.
Left for follow-up (in this bucket but not safe/clear wins here):
- hooks append sequence still re-reads the file per append; the in-memory cache is
cross-process-unsafe and the safe last-line read is fiddly for a low-volume log.
- Fork event re-append/de-dup is nuanced usage-accounting and needs its own change.
* audit batch 18 (agent): count tool-defs in compaction estimate; don't deny an always-allow on persist failure
- maybeCompact's estimate omitted the tool definitions, which ride on every
request, so the real input could exceed the model limit while the message-only
estimate stayed under threshold and compaction never fired (audit medium
compaction.go:58). partitionTools is now computed before maybeCompact (it does
not depend on the messages) and the exposed tools' estimated tokens are added to
both the threshold check and the post-compaction shrink check. + estimateToolDefTokens
and TestEstimateToolDefTokensCountsDefinitions.
- PermissionDecisionAlwaysAllow denied the call when persistPermissionGrant failed,
overriding what the user explicitly allowed (audit medium loop.go:468). Persisting
is now best-effort: a failure (or no sandbox engine) just means the grant is not
remembered and the user is re-prompted next time, not denied now. This also drops
the dead requestEvent.GrantMatched/Grant writes (audit low loop.go:473) — the
emitted event derives them from the sandbox decision via buildPermissionEvent.
Left intentional (documented, not defects): the reactive-retry path omits OnText to
avoid duplicating mid-stream text (loop.go:183); OnContext/MeasureContext is a
complete, tested context-breakdown scaffold awaiting a consumer (types.go:183).
* audit batch 19 (usage): price escalated usage from the event's own model
exec persists payload["model"] on --allow-escalation runs (the model can change
mid-run), but BuildReport priced every event from the session's Metadata.ModelID
and never read the payload, so usage from an escalated model was mis-priced at the
base model (audit medium report.go:17). usageEventPayload now carries Model and
BuildReport prefers it, falling back to the session model when absent (older
events stay byte-identical and behave as before). + TestBuildReportPricesFromEventModelWhenPresent.
* audit batch 20 (sessions): don't copy usage events into a fork (no double-count)
Fork re-appended every parent event, including provider_usage, into the child, so
a usage report aggregating the parent and the fork double-counted the parent's
usage (audit medium store.go:387). Skip EventUsage when copying — it already
counted against the parent and is not part of the conversation history a fork
replays — and record the actual copied count in the fork marker. + TestStoreForkSkipsUsageEvents.
* audit batch 21 (cli/skills): warn on duplicate skill names instead of silently dropping
skills.Load deduplicates same-named skills (first directory wins) and records the
dropped collisions via skills.Duplicates, but nothing surfaced them, so a shadowed
skill just vanished with no signal — and Duplicates had no production caller (audit
low skills.go:91). `skills list` now reports each collision to stderr (keeping
stdout/--json clean), using Duplicates. + TestRunSkillsListWarnsOnDuplicateNames.
Note: skills.Get is kept — it is a by-name accessor used throughout the package's
tests, not dead code.
* audit batch 22 (cli/cron): capture the run error from stream-json stdout
Cron jobs run exec with --output-format stream-json, so a failure is reported as an
`error` event on STDOUT, but the run record's Error was set only from stderr — so a
failed job often recorded an empty/uninformative error while the real message sat
unread in outBuf (audit low cron_run.go:150). On a non-zero exit, scan the stdout
stream-json for the error event's message and prefer it, falling back to stderr.
+ extractStreamJSONError and TestExtractStreamJSONError.
* audit batch 23 (sandbox): richer macOS seatbelt profile (signal + mach-lookup)
The generated sandbox-exec profile was default-deny with only process/sysctl/file
allowances, so two real gaps remained on macOS (sandbox quality report H1):
- A sandboxed script could not signal the children it spawned — `kill` returned
"Operation not permitted" under seatbelt, breaking common `cmd & ... kill`,
test-runner, and timeout patterns. Add `(allow signal (target self) (target pgrp))`,
which stays scoped to the command's own process group.
- XPC to system daemons was denied, so keychain (securityd/trustd), user/group
lookup (opendirectoryd), preferences (cfprefsd), network config
(SystemConfiguration), launch services, and the pasteboard all failed. Add a
curated `(allow mach-lookup ...)` allowlist (`sandboxMachServices`).
Neither touches the file-write or network rules, so the workspace boundary is
unchanged — verified on a real macOS host: `security`/`scutil`/`pbpaste`/self-kill
now succeed while writes to $HOME and /etc are still denied. iokit-open was tried
and dropped as unnecessary. + unit assertions (TestSandboxExecProfileGrantsSignalAndMachLookup)
and extended the darwin integration test with a self-kill case and an
outside-workspace deny check.
* audit batch 24 (sandbox/tools): remove dead OnSandboxDecision hook
RunOptions.OnSandboxDecision had no setter anywhere, yet registry.RunWithOptions
spawned a goroutine (with panic recovery) for it on every sandboxed tool call —
pure per-call overhead for an unreachable callback (sandbox report M1). The sandbox
decision is already surfaced to observers via the permission event
(buildPermissionEvent reads the decision), so this hook is redundant; delete the
field and its call site.
* audit batch 25 (sandbox/cli): exact-path grants — create and revoke by path
The engine already matched ScopeFile grants and DeriveScope already produced them
from path/file tool args, but the CLI could neither create one explicitly nor
revoke a single one (sandbox report L3 / prompt 7). Add:
- GrantStore.RevokePath(tool, path): revokes only the grant whose scope matches the
given file/dir, leaving tool-wide and other-path grants intact (path canonicalized
to absolute like stored scopes). + TestGrantStoreRevokePathRemovesOnlyMatchingScope.
- `zero sandbox grants allow <tool> --path <p>`: persists an exact-file (ScopeFile)
grant, path resolved to absolute so it matches lookups and revokes.
- `zero sandbox grants revoke <tool> --path <p>`: revokes just that grant (default
still revokes all grants for the tool). + TestRunSandboxGrantsCreateAndRevokeByPath.
The TUI permission-card "Allow always (exact path)" option is intentionally left for
the TUI feature branches (it overlaps #187/#188); the engine + CLI now cover the
full create/match/revoke round-trip for file-scoped grants.
* audit batch 26 (sandbox): Engine.Precheck — report violations before execution
Add Engine.Precheck(ctx, request) []Violation, which reports the sandbox
violations that would block a tool request before it runs (sandbox report / prompt 6).
It reuses Evaluate so policy is not duplicated: an allowed or merely-prompted
request yields no violations; a denied request yields its violation (via the shared
violationsFromDecision helper, with a ViolationPolicyDenied fallback for a deny that
carries no structured violation). A nil engine yields nothing.
+ TestEnginePrecheckReportsViolationsBeforeExecution.
Wiring Precheck into the batch-confirmation prompt is intentionally left to the TUI
branch: it touches prompt rendering, and short-circuiting the agent loop's deny path
would skip the before/after-tool hooks that currently fire around a policy-denied
call. The engine API is in place for that consumer.
* audit batch 27 (sandbox): scoped-egress domain prompt with timeout + caching
The scoped-egress proxy failed closed on any host outside the allowlist. Add an
opt-in DomainPrompt callback (prompt 8): an UNKNOWN host (neither allowed nor
explicitly denied) is authorized at request time via the callback instead of being
refused outright. An explicitly denied host is still refused without prompting; the
callback is bounded by PromptTimeout (default 60s, timeout/error => deny); decisions
are cached for the proxy's lifetime so a host is prompted at most once. With no
callback set the strict fail-closed behavior is unchanged. Wired through both the
CONNECT and HTTP handlers via authorizeTarget.
+ TestEgressProxyAuthorizeDomainPrompt / ...TimeoutDenies / ...NilPromptFailsClosed.
The user-facing prompt that supplies the callback is left to the permission UI (TUI)
branch; the egress mechanism (callback + timeout + cache) is in place and tested.
* audit batch 28 (sandbox): AST-based shell analysis (AnalyzeCommand)
Add AnalyzeCommand(script) AnalysisResult, a static shell analyzer built on the
mvdan.cc/sh/v3 parser (prompt 12). Walking the parsed command tree makes it a more
precise second opinion than the regex detector: a program name only counts when it
is an actual command, so `echo "vim ..."`, `printf 'open with vim'`, and
`git commit -m "rm -rf /"` are clean, while real commands are flagged —
interactive (editors/pagers + REPLs, with the same -e/script suppression as
nonInteractiveREPLFlags), destructive (rm -rf/-fr, dd, mkfs, shred, ...), and
network (curl/wget/ssh/nc/...). An unparseable script yields TooComplex so callers
can treat it as higher-risk. + TestAnalyzeCommand (16 cases) / ...EmptyIsClean.
Dependency: mvdan.cc/sh/v3 pinned to v3.11.0 — compatible with the existing go 1.24
directive (the latest v3.13.1 would have bumped the module to go 1.25). It is kept
as a standalone analyzer for callers to consult; feeding it into Engine.Evaluate is
deliberately not done here, since that would change allow/deny classification.
* audit batch 29 (sandbox): unify proxy-env generation (ProxyEnv)
Export ProxyEnv as the single source of truth for proxy-env injection (prompt 9
req 1), replacing the unexported scopedProxyEnv used by the sandboxed-shell path.
+ TestProxyEnv.
Investigation that bounds the rest of prompt 9 (documented in the ProxyEnv doc):
- web_fetch already honors HTTP_PROXY/HTTPS_PROXY — its transport clones
http.DefaultTransport, whose Proxy is http.ProxyFromEnvironment.
- MCP children already inherit the parent env (mergeProcessEnv appends os.Environ),
so they pick up any proxy vars the agent has.
So the only missing link to scope FetchUrl/MCP is a SESSION-level egress proxy whose
address reaches the agent process. That is deliberately NOT wired here: doing it
safely requires allowlisting the active LLM provider's own domain first, or the
agent's provider calls would route through the scoped proxy and be denied — a
correctness pitfall that also can't be verified end-to-end on this host. Flagged
rather than shipped blind.
* audit batch 30 (sandbox): Linux seccomp Unix-socket block — UNVERIFIED, needs Linux CI
Adds a pure-Go (no cgo, no vendored binary) seccomp BPF filter that denies
socket(AF_UNIX) with EPERM on x86-64/arm64, closing the Unix-socket gap bubblewrap's
filesystem/network isolation leaves open (prompt 11):
- internal/sandbox/seccomp.go: the classic-BPF program as a platform-neutral
[]sockFilter, built and UNIT-TESTED (structure + every jump offset in range — the
classic way a hand-written BPF filter goes silently wrong).
- seccomp_linux.go: ApplyUnixSocketBlock installs it via prctl(NO_NEW_PRIVS) +
prctl(SET_SECCOMP, MODE_FILTER). seccomp_other.go is a no-op everywhere else.
- cmd/zero-seccomp: a tiny linux-only exec wrapper that applies the filter then
execs the real command (the standard way to install seccomp in a child with
os/exec); degrades gracefully (warns + runs) if the kernel lacks seccomp.
⚠️ HONEST CAVEAT: I developed this on macOS and CANNOT run the actual AF_UNIX block
here. darwin build, `GOOS=linux` cross-compile, vet, and the structure test all pass,
but the unit test verifies the program's SHAPE, not that the kernel actually blocks
AF_UNIX — that MUST be verified on Linux CI (create a Unix socket under the wrapper
and assert EPERM). Safe to ship as-is: it is linux-only, no-op on every other
platform, and not wired into the bwrap command yet (opt-in prefixing the sandboxed
command with zero-seccomp is the remaining integration), so it cannot regress any
current path.
* audit batch 31 (sandbox): macOS denial log monitor (opt-in, default off)
Re-adds the macOS seatbelt-denial monitor as an opt-in feature (prompt 10). When
Policy.MonitorDenials is set, the sandbox-exec profile tags its default-deny with a
marker and bash tails `log stream` for that tag during the command, appending any
captured denials as a <sandbox_violations> block to the command's stderr so the
model can see what was blocked.
- DenialMonitor (darwin) tails/parses tagged denials with a capped, deduped ring
buffer and noise filtering; no-op stub on other platforms. parseSandboxDenyLine is
unit-tested. + profile-tag test + appendSandboxViolations test.
- Default OFF: with MonitorTag empty, StartDenialMonitor is a no-op and the command
path is byte-for-byte unchanged, so there is no regression or per-command overhead
unless enabled.
CAVEAT: on at least one macOS host (verified live) seatbelt denials are NOT delivered
to the `log stream`-queryable unified log, so the monitor captures nothing there. It
is the standard mechanism and works on macOS versions that do log denials; shipping
it default-off means it never affects whether a command runs.
* audit batch 32 (sandbox): wire seccomp Unix-socket block into bubblewrap (opt-in)
Completes prompt 11: the zero-seccomp helper (which installs the AF_UNIX seccomp
filter) is now actually invoked by the sandbox instead of only existing as a
standalone binary.
- bubblewrapCommandPlan, when Policy.BlockUnixSockets is set, ro-binds the
discovered zero-seccomp helper into the sandbox at its host path and prefixes the
command with it so the filter is installed before the real argv runs.
- findSeccompHelper discovers the helper next to the running executable first, then
on PATH; it returns "" when absent, so the sandbox degrades gracefully (runs
without the extra filter — bubblewrap's fs/net isolation still applies) rather
than failing the command. Exposed as a package var so the wiring is unit-tested
on any OS.
- Reachable from config: SandboxConfig.BlockUnixSockets and .MonitorDenials now map
onto the policy in applyConfiguredSandboxPolicy (opt-in only — config can turn a
hardening flag on, never off). Both default OFF.
- Tests: helper-prefix present when enabled / absent by default; config flags apply
independently and stay off when omitted.
Default OFF and Linux-only effect; the runtime seccomp enforcement still needs a
Linux host/CI to exercise end-to-end (the wiring and BPF program are unit-tested).
* audit batch 33 (sandbox): enforce scoped/deny network policy in web_fetch & web_search
Completes prompt 9: the built-in network tools now honor the SAME allow/deny/scoped
policy the sandboxed shell egress proxy enforces, instead of bypassing it. Before
this, web_fetch/web_search would reach any public host even when the policy was
NetworkDeny or NetworkScoped — the egress proxy only constrained shell commands.
- New shared gate Engine.NetworkHostAllowed(host) reuses the egress domain matcher
(domainAllowed/normalizeDomains/effectiveNetwork) as the single source of truth:
allow → all hosts; deny → none; scoped → only AllowedDomains minus DeniedDomains
(empty allowlist collapses to deny). Disabled policy / nil engine impose nothing.
- web_fetch is now sandbox-aware: the host is checked before the request AND on every
redirect, so a denied/unlisted host fails closed before any dial. Plumbed via a
networkPolicyGate so the no-sandbox path is byte-for-byte unchanged.
- web_search is sandbox-aware: the configured backend's endpoint host is checked
before the query leaves the machine; blocked under deny, and under scoped unless the
backend host is allowlisted.
- The registry already routes sandbox-aware tools through RunWithSandbox when an
engine is present, so this activates automatically wherever the sandbox is wired.
- Tests: engine gate matrix (deny/allow/scoped/denylist/empty/disabled/nil + host:port);
web_fetch & web_search block denied/unlisted hosts without dialing and allow listed
ones; endpoint-host parsing.
* audit batch 34 (sandbox): prove network-policy wiring end-to-end + document opt-in flags
Verification follow-up to batches 31-33; no behaviour change.
- Adds a registry-level test that exercises the FULL chain for the scoped-network
enforcement: registry.RunWithOptions -> engine.Evaluate (passes a scoped,
egress-capable sandbox-exec backend) -> routes web_fetch through RunWithSandbox ->
per-host allowlist blocks an unlisted host before any dial. This proves the
Tool-interface value actually satisfies sandboxAwareTool at the registry boundary,
not just in a direct method call.
- Documents the two opt-in hardening flags (sandbox.blockUnixSockets,
sandbox.monitorDenials) and the scoped web_fetch/web_search behaviour in the README
so the config keys are discoverable. Both remain off by default.
Confirmed wiring: agent loop forwards options.Sandbox to the registry; the registry
routes sandbox-aware tools (web_fetch, web_search) through RunWithSandbox when an
engine is present; MonitorTag is set only on the macOS sandbox-exec plan; the
seccomp helper prefix is added only on the bubblewrap plan when BlockUnixSockets is
set; both config flags map onto the policy via applyConfiguredSandboxPolicy.
* audit batch 35 (sessions): fix Windows CI failure in TestWriteFileSyncRoundTrips
The exact-permission assertion (perm == 0o600) added in batch 17 fails on
windows-latest because Windows does not honor Unix permission bits: Go reports
0o666 for a writable file regardless of the mode passed to OpenFile. This was the
sole cause of the failing Smoke (windows-latest) CI job; macOS and Ubuntu passed.
Guard the perm check with runtime.GOOS != "windows" so it still verifies the mode
on the platforms that enforce it, while the round-trip (write + read-back) is still
asserted on every platform.
* audit batch 36: address CodeRabbit review (5 findings)
- background/process_posix: after SIGKILL, poll until the process group is
actually gone (SIGKILL is async) and return an error if it survives past a
second deadline, instead of returning nil while descendants still race.
- sandbox/safe_command: only treat ')' as a substitution boundary when it closes
an active $(...). A literal ')' (e.g. echo "a) less") no longer splits into a
fake `less` segment and is no longer misflagged as interactive. Depth-tracked so
nested $(...) still isolate their inner command. Direct splitShellSegments test
(fails pre-fix: yields ["echo \"a", "less\""]).
- cron/schedule: the DST fall-back collapse now only applies when `after` sits
exactly on the minute boundary (its "last fire time" form). With sub-minute
precision (e.g. 01:30:30 EDT) the first 01:30 fire already passed, so the repeated
01:30 EST is the legitimate next fire and is returned — restoring strictly-after
semantics. New test fails pre-fix (skipped to next day).
- config/types: maxTurns error message now reads "must be >= 0" to match the check
(0 is accepted; only < 0 is rejected). Negative-maxTurns test asserts the message.
- docs/audit: header date/branch metadata now matches the body (2026-06-13,
fix/audit-batch) instead of the stale 2026-06-11/fix/audit-critical-high snapshot.
Verified: gofmt clean; go vet clean; host + linux + windows builds; affected suites
green. Each behavioural fix ships with a test confirmed to fail without the change.
* audit batch 37: address CodeRabbit re-review (12 findings)
Findings on the audit code from batches 21-36; each behavioural fix ships with a
regression test (several confirmed to fail without the change).
sandbox/analyzer (AnalyzeCommand):
- Unwrap launcher prefixes (sudo/env/nice/bash -c/...) and recurse into `sh -c`
payloads so `sudo rm -rf`, `env curl …`, `bash -c 'vim x'` classify on the real
command, not the launcher. Mirrors DetectInteractiveCommand.
- Honor rm's `--` end-of-options marker so `rm -- -rf` (delete a file named -rf) is
not flagged destructive.
sandbox/engine: NetworkHostAllowed now shares one backend-aware effective-mode
helper with Evaluate, so scoped collapses to deny when the backend can't enforce
scoped egress — the per-tool gate can no longer permit traffic the engine-level
decision would fail closed.
sandbox/egress: cache domain-prompt decisions by host:port, not host, so one
"allow" can't authorize other ports on the same host.
sandbox/safe_command: the segment splitter is now escape-aware (\| , \" ) so an
escaped operator/quote can't manufacture a fake interactive segment.
sandbox/log_monitor (darwin): Wait() the `log stream` child after draining stdout
so the cancelled process isn't leaked as a zombie.
tools/web_search: confine the search backend to same-host redirects (with a limit)
so a redirect chain can't egress to a host the scoped/deny policy never authorized.
cli/sandbox: an explicit empty --path (`--path=` / `--path ""`) now fails closed
instead of silently widening an allow to tool-wide or a revoke to all-grants.
cli/skills: warn on stderr when duplicate-skill detection fails instead of
swallowing the error.
specialist/exec: don't mask a failed orphan-kill — join the TerminateProcess error
with the SetPID error so a surviving process is visible.
background/process_posix: resolve the PGID and only group-signal when pid leads its
own group; otherwise signal the individual PID. Avoids both a false-success on a
non-leader ESRCH and the hazard of signalling an unrelated (possibly our own) group.
Verified: gofmt + go vet clean; host + linux + windows builds; full go test ./...
green.
* audit batch 38: address CodeRabbit re-review round 2 (3 findings)
Follow-ups to batch 37, each with a regression test:
- safe_command (splitter): a ')' is now only a substitution boundary when it
closes an active, UNQUOTED $(...). A quote-state stack saved at each '$(' lets a
substitution run in a fresh quoting context and restores the outer quoting on
its matching ')'. Fixes `echo $(printf "a) less")` splitting into a fake `less`
segment, while still isolating the inner command of `echo "$(vim x)"`.
- egress: deduplicate in-flight domain prompts per host:port. Concurrent requests
for the same unknown target now wait on the first prompt and reuse its cached
decision instead of each firing a prompt (a prompt storm) and racing to write
the final decision. Race-tested with 25 concurrent authorizers → one prompt.
- analyzer + safe_command: wrapper option-value consumption is now wrapper-specific
(wrapperValueOptionsByProg) instead of a global option set. A valueless flag like
`sudo -n` no longer swallows the real payload, so `sudo -n rm -rf` /
`sudo -n curl …` are correctly classified destructive/network; `sudo -u root vim`
still consumes its value. Shared by the AST analyzer and the regex detector so
they can't diverge.
Verified: gofmt + go vet clean; host + linux + windows builds; go test ./... and
-race on sandbox/background all green.
* audit batch 39: address CodeRabbit re-review round 3 (12 findings)
All verified against current code; each behavioural fix has a regression test.
Security / correctness:
- tools/web_search: redirect policy now also refuses a scheme change, blocking a
same-host https→http downgrade that would send the query + bearer token over
plaintext.
- sandbox/safe_command: a ')' is only a substitution boundary when it closes an
active UNQUOTED $(...), and backticks now get the same quote-state save/restore
as $( — so `echo "`a | less`"` and `echo "$(true | less)"` no longer hide the
inner pager from interactive detection.
- sandbox/analyzer + safe_command: wrapper option-value consumption stays
wrapper-specific, and unwrapping no longer aborts on a dynamic ($x) wrapper arg
once a wrapper is active, so `env "$opts" curl` / `sudo "$x" rm -rf` are still
classified.
- sandbox/egress: in-flight prompts deduped per host:port (carried from batch 38);
the prompt callback is now cancellable (context cancelled on timeout) so a stuck
prompt can abort instead of leaking; signature threaded through.
- sandbox/runner: per-plan unique denial tag (pid+counter) so concurrent monitored
sandbox-exec runs can't ingest each other's denials; helper selection now
requires an executable regular file so a non-exec sibling degrades gracefully.
- mcp/network_client: aggregate SSE event-size cap (not just per-line) so a server
can't grow memory with many unterminated data: lines.
- specialist/accounting: a previously-recorded stop/usage with an empty runId is a
catch-all and now suppresses a later runId-bearing duplicate.
- sessions/lineage: Tree fetches the root via Get() first so a corrupt root
surfaces its real error instead of degrading to "not found".
Docs/tests:
- background/terminate: docstring now states the leader-only group-kill (PID-only
fallback for non-leaders); test child-liveness probe uses ESRCH specifically.
- cli/sandbox: grant usage strings include --path.
Verified: gofmt + go vet clean; host + linux + windows builds; go test ./... and
-race on sandbox/background all green.
* audit batch 40: address CodeRabbit re-review round 4 (8 fixed, 1 skipped)
Verified each against current code.
Fixed:
- sessions/lineage: PANIC FIX — store.Get returns (nil, nil) for a missing
session, so the batch-39 root prefetch could nil-deref. Guard root == nil and
return a not-found error. Regression test added (was a panic).
- agent/compaction: the reactive recover() now records lowWaterMark in the SAME
combined (messages + tool-defs) token domain maybeCompact uses, so the proactive
shrink-guard compares like with like. recover() takes tools; callers pass
request.Tools.
- mcp/network_client: SSE aggregate counter no longer counts a joining newline for
the first data line (only when prior content exists), avoiding premature rejects.
- sessions/store: fsync the parent directory after the metadata rename so the
rename is durable across a crash (no-op on Windows, which can't fsync a dir).
- sandbox/egress: parse the target port with strconv.Atoi instead of
net.LookupPort (no context-unaware resolver use).
- sandbox/runner: escape the per-plan denial tag through sandboxProfileString when
embedding it in the seatbelt profile.
- sandbox/seccomp_linux: guard against an empty filter program before taking
&kernelFilters[0] (defensive; the program is fixed non-empty today).
- cli/cron_run_test: assert last-error-wins with two error events.
- background/terminate, cli/sandbox usage, process_posix_test ESRCH (carried).
Skipped (with reason):
- zerogit worktree-rename (Y='R') test: git porcelain v1 -z only emits the
rename-source field when the INDEX column is R/C; the worktree column never
carries 'R' with a source field in v1, so the fixture would test output git
doesn't produce, and making parseStatus consume on code[1]=='R' would risk
mis-consuming the following entry for a plain ` R` status. Left as-is.
Verified: gofmt + go vet clean; host + linux + windows builds; go test ./... and
-race on sandbox/background all green.
* audit batch 41: address CodeRabbit re-review round 5 (4 findings)
All verified against current code; code fixes have regression tests.
- agent/compaction: summarizeWithFallback's reduce pass no longer swallows ALL
errors. A context-limit error still falls back to the joined partial summaries
(documented "extreme" case), but a non-context failure (auth/network/provider)
now surfaces unchanged, per the function's contract. Two tests (propagate vs.
joined-fallback); the propagate test fails without the fix.
- sandbox/safe_command + analyzer: wrapper value-consuming options now include the
long spellings (sudo --user root, env --unset NAME, timeout --signal …), so a
long flag's value is no longer mis-read as the program (interactive/destructive/
network detection bypass). A `--flag=value` token carries its own value and never
consumes the next token. Tests for space and = forms in both the AST analyzer and
the regex detector.
- docs/audit: the status-ledger summary table's open/partial cells (64/14) were the
lone outlier — the ledger's own narrative and 2026-06-13-reverification.md both
state 66/12. Corrected to 66/12 (66+12+97 = 175); derivation: body 69 open + 18
partial minus the 9 already-fixed-on-main (3 open + 6 partial) = 66/12 = 78
pending, fixed 88+9 = 97. Added a comment documenting the summary-vs-body
relationship. The reverification doc already states 66/12, so both now agree.
Verified: gofmt + go vet clean; host + linux + windows builds; full go test ./...
green.
* audit batch 42: address CodeRabbit re-review round 6 (1 fixed, 1 kept w/ proof)
CodeRabbit re-surfaced two "duplicate" refinements; CI was already green (its 6th
pass was a COMMENT with no new findings).
- cli/sandbox_test (TestRunSandboxGrantsRejectsEmptyPath): now seeds a tool-wide
grant and asserts grant-store immutability after EACH rejected empty-`--path`
call (reflect.DeepEqual before/after), so a mutation in one call can't be masked
by a later one — and a buggy "revoke all" is actually observable (it's a no-op on
an empty store). Stronger than the suggested patch.
- zerogit.parseStatus: KEPT consuming the rename source on the index column only
(code[0]), with an explanatory comment. Verified empirically that git porcelain
v1 -z reports a rename/copy (and emits the extra source field) only in the index
column: `git mv` → "R new\0old\0"; a worktree-only rename → " D old\0?? new\0"
(delete + untracked, never "R" in code[1]). Consuming on code[1]=='R'/'C' would
match no real git output and would only risk mis-consuming the next entry on
malformed input, so the suggestion is declined as incorrect for v1.
Verified: gofmt + go vet clean; host + linux + windows builds; affected suites green.
|
||
|
KRATOS
|
ae4ce8aae2 |
wedge build: web_search + LSP client/diagnostics (stages 01-03) (#175)
* stage 01: web_search tool Adds a web_search tool alongside web_fetch: discovers URLs via a pluggable searchBackend (env-configured generic JSON backend; swappable for Exa/You/ Tavily/Brave) and formats ranked results as a numbered list. Registered in CoreNetworkTools(). Deviation from the stage spec: the spec said readOnlySafety, but the repo enforces (TestCoreNetworkToolsExposePromptMetadata) that every core network tool is prompt-gated network egress — so web_search matches web_fetch's posture instead of being auto-allowed. Necessary fallout from adding the tool: knownToolNames (+web_search), the network-tools metadata test (now covers both), and a stale 'unknown tool' fixture that had used web_search. * stage 02: LSP client foundation New internal/lsp package: a direct (self-spawned, stdio) Language Server Protocol client so unattended runs see real compiler diagnostics without an open editor. - client.go: JSON-RPC 2.0 with LSP Content-Length framing; concurrent Call/Notify, id-matched responses, server->client request auto-reply, notification handler, context cancellation, read-loop teardown. - server.go: process lifecycle (spawn via os/exec, initialize/initialized handshake, graceful shutdown+exit with kill fallback). Handshake factored out (performInitialize) for pipe-based testing. - registry.go: extension -> server command (.go/.ts/.tsx/.js/.jsx/.py/.rs), languageId map, PATH availability via exec.LookPath. - types.go: minimal LSP types + PathToURI/URIToPath (Windows-drive aware). Stdlib-only; tested over in-memory pipes with a stub server (no real gopls), including -race for the concurrent id router. * stage 03: LSP document sync + diagnostics Makes the LSP client produce actionable signal so ZERO can ask 'did my edit compile?' headlessly. - documents.go: per-server session with document lifecycle (didOpen/didChange full-sync/didClose), a URI->diagnostics store fed by publishDiagnostics, and WaitForDiagnostics with a publish-baseline + 300ms quiet debounce (servers never signal 'analysis complete'). - diagnostics.go: FormatDiagnostics (path:line:col: severity: message, 1-based), HasErrors, FilterBySeverity, severity labels. - manager.go: Manager — one lazily-started, reused server per language (keyed by binary), routes a file to the right server, Check(ctx,path,text) -> settled diagnostics, graceful Shutdown; unknown extension -> (nil,nil), not an error. Concurrency-safe; tested under -race over in-memory pipes with a stub server (no real gopls). FormatDiagnostics takes the path explicitly since a Diagnostic carries only a range, not its document. * wedge: address review on #175 (LSP robustness + web_search provider) - lsp/client.go: Close() now disables the client — Call/Notify reject after close instead of leaking a pending entry / writing to a dead conn. - lsp/documents.go: serialize same-document syncs through the Notify write via a per-URI lock and commit version/open only on success; drop delayed publishDiagnostics for a superseded (older positive) version; WaitForDiagnostics now reports whether a fresh publish arrived. - lsp/manager.go: Check returns (nil,nil) when no publish newer than baseline arrives (don't return a stale prior result for new text); Shutdown joins and returns per-server shutdown errors instead of always reporting success. - lsp/diagnostics.go: FormatDiagnostics collapses embedded whitespace so each diagnostic stays one physical line. - tools/web_search.go: forward ZERO_WEBSEARCH_PROVIDER in the request so the config knob isn't inert. Adds tests: close-disables-client, stale-version drop, multi-line message collapse, Shutdown error propagation, httptest backend (provider+parse). * wedge: degrade gracefully when an LSP server binary is missing (#175 review) Manager.Check translated a missing-binary start error (exec.ErrNotFound) into (nil,nil) so a configured extension on a machine without gopls/ pyright/rust-analyzer degrades exactly like an unsupported extension — LSP diagnostics are an opportunistic layer, not a hard dependency. Other start failures still surface. Adds a test injecting an exec.ErrNotFound starter. * wedge stage 04: self-correct loop (#176) * stage 04: self-correct loop After a successful mutating tool call, verify the changed files (LSP diagnostics + project tests) and feed failures back to the model to fix — bounded and autonomy-gated, so an unattended run fixes its own mistakes without looping forever. - internal/agent/selfcorrect.go: SelfCorrector with injectable diagnosticsChecker (lsp.Manager adapter) and projectVerifier (verify.DetectPlan+Run); AfterEdit produces a redacted, model-facing corrective message; per-run attempt ceiling; low autonomy reports without auto-attempting; medium/high auto-correct. - internal/agent/loop.go: invoke AfterEdit after a successful mutating tool call (ChangedFiles non-empty), appending feedback to messages. nil SelfCorrect (the default) is a no-op — read-only tools never trigger it. - internal/agent/types.go: add Options.SelfCorrect *SelfCorrector (the explicit options-struct change). Single-pass verify per call (the agent loop + model edits provide the multi- attempt cycle, not selfverify's same-command re-runs). Tested with fakes (no gopls/shell): fail-then-pass, max-attempts abort, read-only skip, low-autonomy report-only, LSP-only failure, secret redaction; -race clean. CLI activation (a --self-correct flag building the SelfCorrector in exec.go) is intentionally left out to honor this stage's file scope — the engine is exported and ready to wire. * wedge: propagate file-read error in self-correct LSP checker (#176 review) CodeRabbit/golangci-lint (nilerr) flagged lspDiagnosticsChecker.Check returning (nil, nil) after os.ReadFile fails. Propagate the error instead; the sole caller (inspect) already skips files whose Check errors, so a deleted/unreadable file still degrades quietly while future callers can decide. Adds a regression test. * wedge: surface non-degradable verify errors in self-correct (#175 review) SelfCorrector.inspect silently continued on every checker.Check error and ignored verifier errors, so a real LSP startup/sync failure, an unreadable changed file, or a failed plan detection produced report.Failed=false -> OutcomePassed with no signal. Manager.Check already degrades missing servers to (nil,nil) and DetectPlan returns an empty plan (not an error) when nothing is detected, so a non-nil error from either is a genuine failure. Record such errors in CorrectionReport.InspectErrors, fail the pass, and surface them in the model feedback. Adds regression tests for both paths. * wedge: drop third-party search-provider names from web_search comments Keep code/comments provider-neutral: describe the backend interface as 'any hosted search API' rather than naming specific services. * wedge: defer self-correct feedback until after the tool-call batch (#175 review) Self-correct feedback was appended as a user message inside the per-tool-call loop, so a multi-tool-call turn whose first (mutating) call failed verification produced: assistant tool_use batch -> tool_result 1 -> user feedback -> tool_result 2. A user message interleaved between tool_results breaks strict provider replay and later session replay/compaction. Accumulate feedback during the batch and append it once after the loop (same after-batch pattern already used for mid-run model escalation), so every advertised tool call keeps its tool_result contiguous. Adds a two-tool-call regression test asserting the feedback lands after both tool_results. * wedge: batch self-correct AfterEdit once per turn over union of changes (#175 review) Deferring only the feedback emission still ran SelfCorrect.AfterEdit once per mutating tool call, which (a) consumes the per-run attempt budget multiple times in a single turn and (b) verifies an intermediate edit that a later call in the same batch already superseded — AfterEdit is documented as one per-run instance holding attempt state, so this is a correctness issue, not just transcript shape. Aggregate every mutating tool's ChangedFiles during the batch and call AfterEdit once afterward over the deduped union, appending its feedback once (still after all tool_results, preserving contiguity). Adds dedupeStrings and a regression test asserting two writes in one turn consume exactly one attempt and emit one feedback message. |
||
|
KRATOS
|
06ea4f004b |
Edit safety: file read/write tracking, external-modification detection, post-apply hooks (#172)
* Edit safety: file version tracking, conflict detection, hook wiring
Phase 1 — FileTracker (internal/tools/file_tracker.go): per-session
map[absPath]{sha256, size, mtime}; CheckConflict flags drift only against
a known baseline, so first-touch writes are unaffected. nil = no-op.
Phase 2 — external-modification detection: read_file records the
whole-file baseline; edit_file and write_file --overwrite refuse to
clobber a file whose on-disk content drifted from it and tell the model
to re-read first, then re-baseline to what they wrote. apply_patch
forgets the files it rewrote (git apply has its own context guard).
Threaded via RunOptions/agent.Options FileTracker, built once per session
in the exec, spec, and TUI paths.
Phase 3 — post-apply validation via hooks: the dormant hooks.Dispatcher
is wired into executeToolCall. beforeTool hooks can veto a call
(DenialHookBlocked); afterTool hook output is scrubbed and surfaced to
the model. DispatchOutcome now collects hook stdout/stderr. Per-session
dispatcher built from merged user+project hooks.json; fails open and runs
nothing until a hook is configured.
* edit-safety: address review — fail closed on read, redact hook reason
- write_file overwrite: if a tracked file cannot be re-read to verify it,
fail CLOSED (refuse the overwrite) instead of silently proceeding — the
read error previously skipped the conflict check and could clobber an
externally modified or replaced file.
- blockedByHookResult: scrub outcome.Reason through redaction before placing
it in model-visible ToolResult.Output; this intercepted path bypassed the
registry's normal output-redaction boundary.
- TestHashContentIsStableAndDistinguishing: compare separately stored hashes
so the stability check is not a same-expression comparison (SA4000).
* edit-safety: flag ToolResult.Redacted when hook output is scrubbed
appendHookFeedback now reports whether scrubbing changed the afterTool
feedback, and blockedByHookResult reports whether it scrubbed the reason;
both set ToolResult.Redacted so the hook-intercepted paths match the
registry's redaction contract. Adds tests asserting a secret-bearing hook
reason/feedback is scrubbed and flips Redacted.
|
||
|
KRATOS
|
3f3de1a29b |
permissions: scope-enforced grants (permission-UX Phase 0 + 1) (#170)
* permissions: show grant scope on the permission card and decided rows The permission card only named the tool, so 'always' read as a blind tool-wide yes. Derive a concise scope from the call's args (the file path, directory, or working dir it touches) and show it on the card and the persisted decision row, so a user can see exactly what an allow covers. Phase 0 of the permission-UX work: presentation only, no storage change. permissionScope + scope plumbing through PermissionEvent/PermissionRequest; covered by TestPermissionScope. * permissions: spec for Phase 1 scope-enforced grants * permissions: enforce grant scope so "always allow" covers only what the card showed Phase 0 displayed the scope a tool call touches but the persisted grant was still tool-wide: an "always allow" on a write to one file silently authorized every write. This scopes the grant to exactly that file or directory. sandbox/scope.go centralizes scope derivation (DeriveScope), absolute-path resolution (resolveScopeAbs, anchored to the workspace so a grant never leaks across projects), and matching (grantCovers: file = exact, dir = subtree, empty = tool-wide; a tool-wide request is never covered by a narrower grant). Grants are now stored per-tool as a list (schema v2, with v1 files migrated as tool-wide grants). Lookup takes the request's absolute scope and applies deny-wins / most-specific-allow precedence. engine.Grant anchors a relative scope to the workspace; engine.Decide derives and matches the request scope. agent.persistPermissionGrant forwards the scope, and permissionScope now shares DeriveScope so the card and the stored grant can never diverge. Canonicalizing tool keys on read also closes the whitespace-padded-key lookup miss. Covered by sandbox scope/grant/engine tests and the agent/TUI always-allow tests. * sandbox: fix Windows-only TestResolveScopeAbs by using truly-absolute paths A leading separator (\proj\a) is rooted but not absolute on Windows (filepath.IsAbs requires a volume like C:), so the absolute-passthrough case was treated as relative and anchored onto the workspace root, producing a doubled path. Build the root via filepath.Abs so the test exercises a genuinely-absolute path on every platform. * test: isolate session store in exec/tui tests so go test stops writing to $HOME TestRunExecStreamJSONRunStartUsesResolvedAPIModel, TestRunExecReadsStreamJSONPromptFromStdin, and TestPromptSubmitInjectsLiveSessionModelContext ran real exec sessions but only isolated cwd (t.TempDir + getwd), not the session store, so the default store resolved sessions.DefaultRoot() off the real $HOME and persisted into ~/.local/share/zero/sessions on every run. Set XDG_DATA_HOME to a per-test temp dir (the existing convention in exec_test.go / exec_scope_test.go) so each run is fully isolated. * permissions: address review — clean root-equiv scopes, label+fit scope rows - DeriveScope: filepath.Clean the path before the root check so ./, ./., and a/.. collapse to the workspace root and surface as tool-wide instead of a narrower directory grant (which re-prompted inconsistently). Adds regression cases. - renderPermissionRow: prefix the scope segment as scope:<value> across all three branches and fitStyledLine the allow branch (it returned unfit and could overflow narrow terminals). - TestGrantReplacesSameScope: assert the setup store.Grant errors so a setup failure fails at the real cause, not a stale-state assertion. * docs: point permission-scope spec at grant_scope.go The design spec named internal/sandbox/scope.go as the source of truth, but DeriveScope/resolveScopeAbs/grantCovers live in grant_scope.go (scope.go is the unrelated write-roots Scope from #162). Fix the heading and the test reference (grant_scope_test). |
||
|
gnanam1990
|
44411578e9 |
providers/anthropic: forward reasoning effort as extended thinking
Map a requested reasoning effort to an Anthropic thinking budget and enable extended thinking, raising max_tokens so the budget plus a response both fit. Crucially, capture the streamed thinking / redacted_thinking blocks (text + signature) and preserve them on the assistant message so they are replayed verbatim at the start of the next turn — without this, Anthropic rejects every tool-using follow-up. Adds zeroruntime carriers (Message.Reasoning, ToolCall. Signature, ReasoningBlock) shared with the upcoming Gemini work. No effort = no thinking, so default runs are unchanged. |