docs(contributing): add contribution volume and batching guidance

Documents expectations for keeping contribution flow sustainable during the
stabilization phase: one issue per bug (no bundled multi-finding audits), keep
only a few items open at a time, and hold AI-assisted contributions to the same
per-item verification bar. Complements the existing issue-approved gate, which
already covers PR-to-issue linkage and scope; the gap was submission volume.

.github/ISSUE_TEMPLATE/config.yml

@@ -1,5 +1,11 @@
 blank_issues_enabled: false
 contact_links:
+  - name: Questions and setup help
+    url: https://github.com/Gitlawb/zero/discussions/categories/q-a
+    about: Ask in Discussions Q&A. Include zero --version, your OS, provider/model, and zero doctor output for setup problems.
+  - name: Feature ideas
+    url: https://github.com/Gitlawb/zero/discussions/categories/ideas
+    about: Propose and discuss ideas in Discussions before opening a PR (Zero is in controlled development, see CONTRIBUTING.md).
   - name: Security vulnerability
     url: https://github.com/Gitlawb/zero/security/advisories/new
     about: Report security issues privately via GitHub Security Advisories (see SECURITY.md). Do not file a public issue.

.github/ISSUE_TEMPLATE/feature_request.yml

@@ -1,12 +1,22 @@
-name: Feature request
-description: Propose a new capability or change
+name: Feature request (previously discussed)
+description: File only after your idea gained traction in Discussions Ideas or a maintainer asked for a tracking issue
 labels: ["enhancement"]
 body:
   - type: markdown
     attributes:
       value: |
-        Please do not begin implementation until the issue has been reviewed and marked with the
+        New feature ideas start in [Discussions Ideas](https://github.com/Gitlawb/zero/discussions/categories/ideas),
+        not here. This form is for the step after that: an idea that gained traction in Discussions,
+        or one a maintainer asked to have tracked as an issue. Please link the discussion below.
+        Do not begin implementation until the issue has been reviewed and marked with the
         `issue-approved` label (see CONTRIBUTING.md).
+  - type: input
+    id: discussion
+    attributes:
+      label: Ideas discussion
+      description: Link to the Discussions Ideas thread (or note which maintainer asked for this issue).
+    validations:
+      required: true
   - type: textarea
     id: problem
     attributes:

.gitignore

@@ -55,3 +55,6 @@ docs/superpowers/
 zero-termux-arm64
 zero-linux-sandbox
 zero-seccomp
+
+# Generated benchmark reports are evidence for one configuration, never tree state
+internal/perfbench/reports/*.json

AGENTS.md

@@ -1,27 +1,71 @@
----
-description: Project guidelines for the open-source CLI coding agent (zero) repository.
-globs: "*.go, *.js, *.md, *.json, *.toml, *.yaml, *.yml"
-alwaysApply: false
----
+# Repository Guidelines for Zero
 
-# Repository Conventions for Zero
+These instructions apply to all work in this repository. For the user-facing
+guide to extending Zero with specialists, hooks, plugins, MCP, and skills, see
+[docs/EXTENDING.md](docs/EXTENDING.md).
 
-This file outlines the project conventions and repository guidelines for coding agents when working on the `zero` repository. For the general guide on how to extend Zero (write specialist sub-agents, hooks, plugins, MCP, skills), see [docs/EXTENDING.md](docs/EXTENDING.md).
+## 1. Contribution and Pull Request Rules
 
-## 1. Project Conventions
+- Before opening any pull request, **all contributors**—including maintainers,
+  community contributors, and coding agents—must read and follow
+  [CONTRIBUTING.md](CONTRIBUTING.md).
+- Community pull requests require an existing parent issue with the
+  `issue-approved` label. Team members may open pull requests through the
+  internal development process described in `CONTRIBUTING.md`.
+- Keep each change focused on the approved or assigned scope. Do not include
+  unrelated fixes, refactors, formatting churn, generated output, or existing
+  local changes in the same commit or pull request.
+- Discuss new implementation languages, runtimes, major dependency changes,
+  and broad architectural rewrites with maintainers before implementation.
+- Pull request descriptions must explain what changed and why, link the parent
+  issue when required, and list the tests or verification performed. Include
+  screenshots or a short recording for user-visible UI changes when practical.
 
-- Build with `make`, not `go build` directly.
-- Tests live next to the source file (`foo_test.go` next to `foo.go`).
-- Run `make lint` before opening a PR.
-- Never edit files under `third_party/` — those are vendored.
-- Unify functions/methods where possible to prevent codebase inflation. Prefer a single cross-platform function with conditional checks over duplicating helpers per platform.
+## 2. Repository and Implementation Conventions
 
-## 2. Guidelines for Coding Agents
+- Use the Go version declared in `go.mod`. Do not hardcode a different local
+  toolchain version in scripts or documentation.
+- Use the repository build and release commands (`make` and
+  `go run ./cmd/zero-release ...`) instead of inventing parallel build flows.
+- Keep tests beside their source files (`foo_test.go` next to `foo.go`). Add a
+  regression test for behavior changes and run affected concurrent code under
+  the race detector.
+- Never edit files under `third_party/`; they are vendored.
+- Prefer one cross-platform function with small conditional checks over
+  duplicated platform-specific helpers when the behavior can remain unified.
+- Do not commit generated benchmark reports from
+  `internal/perfbench/reports/*.json`; reports are configuration-specific
+  evidence, not repository state.
+- Preserve the user's working tree. Do not overwrite, delete, stage, or commit
+  unrelated tracked or untracked files.
 
-If you are an AI coding agent executing tasks in this repository, you **MUST** run all Go code quality and security checks before committing code or completing your task:
+## 3. Required Validation
 
-1. **Format & Vet**: Run `go fmt ./...` and `go vet ./...`.
-2. **Lint**: Run `go run github.com/golangci/golangci-lint/v2/cmd/[email protected] run --enable-only unused,ineffassign,staticcheck ./...`.
-3. **Security**: Run `go run golang.org/x/vuln/cmd/[email protected] ./...`.
+Run validation from the repository root before committing, opening a pull
+request, or completing an implementation task:
 
-If any check fails or cannot be run, do not ignore it. Prompt the user for instructions or setup assistance before proceeding.
+1. **Formatting check**: `make fmt-check`. If it fails, format with
+   `go fmt ./...` (or `make fmt`) and run the check again.
+2. **Vet**: `go vet ./...` (or `make vet`).
+3. **Tests**: `go test ./...`. Use `make test` for the full race-enabled suite,
+   or run focused tests with `-race`, when concurrency is affected.
+4. **Build**: `go run ./cmd/zero-release build`.
+5. **Smoke test**: `go run ./cmd/zero-release smoke`.
+6. **Advisory lint**:
+   `go run github.com/golangci/golangci-lint/v2/cmd/[email protected] run --enable-only unused,ineffassign,staticcheck ./...`.
+7. **Security**:
+   `go run golang.org/x/vuln/cmd/[email protected] ./...`.
+8. **Diff hygiene**: `git diff HEAD --check` (covers staged and unstaged
+   tracked changes).
+
+`make lint` currently runs the formatting check and `go vet`; it does **not**
+run golangci-lint. The pinned golangci-lint job is advisory in CI while the
+existing repository-wide backlog is cleaned up. Fix findings introduced by or
+related to the current change. Unrelated pre-existing advisory findings do not
+justify expanding a focused pull request; report them separately.
+
+Formatting, vet, tests, build, smoke, diff hygiene, and govulncheck are hard
+requirements. If a related check fails, fix it. If a required check cannot run
+because of the environment or fails for an unrelated external reason, report
+the exact failure and obtain maintainer direction rather than silently ignoring
+it.

CONTRIBUTING.md

@@ -8,6 +8,13 @@ stabilizes. Community participation is still welcome, especially through bug
 reports, feature requests, reproduction details, logs, and discussion on existing
 issues.
 
+Questions and setup help belong in
+[GitHub Discussions](https://github.com/Gitlawb/zero/discussions) (see
+[SUPPORT.md](SUPPORT.md)), not in the issue tracker. Feature ideas start in the
+[Ideas](https://github.com/Gitlawb/zero/discussions/categories/ideas) category;
+an idea that gains traction there is the path to an approved issue and,
+eventually, an accepted PR.
+
 Please read this policy before opening a pull request. Pull requests that do not
 follow it may be closed without review.
 
@@ -33,7 +40,11 @@ unless the issue has been explicitly approved for outside contribution.
 There are three primary ways to contribute:
 
 1. Open a bug report with clear reproduction steps.
-2. Open a feature request describing the problem, use case, and expected behavior.
+2. Propose a feature in
+   [Discussions Ideas](https://github.com/Gitlawb/zero/discussions/categories/ideas),
+   describing the problem, use case, and expected behavior. A feature-request
+   issue is filed only after the idea gains traction there or a maintainer asks
+   for one.
 3. Add useful context to an existing issue, such as logs, screenshots, examples,
    affected versions, or reproduction cases.
 
@@ -46,7 +57,8 @@ All community pull requests require an approved parent issue.
 
 Before starting implementation work:
 
-1. Open an issue using the appropriate issue template.
+1. For a bug, open an issue using the bug-report template. For a feature, start
+   in Discussions Ideas; the feature-request issue comes after the discussion.
 2. Describe the bug, feature request, or functional change clearly.
 3. Wait for the core team to review the issue.
 4. Do not open a pull request unless the issue has the `issue-approved` label.
@@ -55,6 +67,28 @@ An approved issue means only that the specific issue has been accepted for work.
 It does not approve unrelated future work, future pull requests, or future issues
 from the same author. Each issue is reviewed case by case.
 
+## Contribution Volume and Batching
+
+To keep triage and review sustainable while the project stabilizes, please keep
+your contribution flow to a size the core team can realistically work through:
+
+- One issue per bug. File each defect as its own issue with its own
+  reproduction. Please do not bundle several unrelated findings into a single
+  "audit" issue; separate issues are far easier to triage, approve, and close
+  individually.
+- Keep only a few items open at a time. If you already have several issues or
+  pull requests awaiting review, let those move through the queue before opening
+  more. A large simultaneous burst from one contributor slows review for
+  everyone, even when each item is valid.
+- AI-assisted contributions are welcome, held to the same bar. Automated audits
+  and generated patches are fine, but each issue and pull request must be
+  individually verified, scoped, and reproducible before you open it. Please do
+  not post bulk machine-generated findings unfiltered.
+
+These apply to all contributors and are about review bandwidth, not the value of
+any single report. Maintainers may ask you to consolidate, split, or hold
+submissions to keep the queue manageable.
+
 ## Pull Request Requirements
 
 Before opening a community pull request, make sure all of the following are true:
@@ -133,7 +167,12 @@ Clear bug reports are the most useful way to help the project during this phase.
 
 ## Submitting a Feature Request
 
-When filing a feature request, please include:
+Feature requests start as a thread in
+[Discussions Ideas](https://github.com/Gitlawb/zero/discussions/categories/ideas),
+not as an issue. The feature-request issue form exists for the step AFTER that:
+when an idea has gained traction in Discussions or a maintainer has asked for an
+issue to track it. Whether in the discussion or the eventual issue, please
+include:
 
 - The problem you are trying to solve.
 - The behavior or functionality you are requesting.
@@ -156,5 +195,9 @@ contribution phase.
 
 ## Getting Help
 
-If you are unsure whether a change is appropriate, open an issue first and ask.
-Please keep discussion focused, specific, and respectful of maintainer time.
+General questions, setup help, and "is this a bug or my config" uncertainty all
+go to
+[Discussions Q&A](https://github.com/Gitlawb/zero/discussions/categories/q-a).
+If you are unsure whether a proposed code change is appropriate, raise it in
+Discussions Ideas or on the relevant issue first and ask. Please keep discussion
+focused, specific, and respectful of maintainer time.

README.md

@@ -195,7 +195,7 @@ Useful controls:
 | `Ctrl+P` / `Ctrl+N` | previous / next item in menus (arrows still work) |
 | `Shift+Tab` | cycle permission mode |
 | `Ctrl+B` | show/hide the sidebar |
-| `Ctrl+C` | cancel or exit |
+| `Ctrl+C` | cancel, exit, or return from a `/btw` conversation |
 
 Common slash commands:
 
@@ -205,6 +205,7 @@ Common slash commands:
 | `/spec`, `/plan` | draft and review a plan before building |
 | `/image` | attach an image for vision-capable models |
 | `/resume`, `/rewind` | continue or roll back local sessions |
+| `/btw [question]` | ask in an isolated fork without adding the side conversation to the main session |
 | `/loop` | repeat a prompt or custom `/command` on an interval (`/loop 5m /babysit-prs`) or self-paced |
 | `/compact`, `/context` | manage context usage |
 | `/permissions`, `/tools` | inspect available tools and policy |
@@ -358,6 +359,12 @@ go run ./cmd/zero-release smoke
 go run ./cmd/zero-perf-bench
 ```
 
+Experimental: `ZERO_OPENAI_TURN_SESSION=1` enables the optimized OpenAI turn
+session (background connection prewarm + request-prefix telemetry) for headless
+`zero exec` runs against official OpenAI profiles. Off by default; `0`/`false`
+disable. A/B-benchmark it by running the same `zero-perf-bench` suite with the
+variable unset and set.
+
 ### Code Quality and Security Checks
 
 Before committing any changes, ensure all Go code quality and security checks pass. Pinned `go run` commands matching CI constraints can be used directly without prior installation:
@@ -377,6 +384,15 @@ go install github.com/golangci/golangci-lint/v2/cmd/[email protected]
 go install golang.org/x/vuln/cmd/[email protected]
 ```
 
+The installed binaries land in `$GOBIN` when it is set, otherwise in
+`$GOPATH/bin` (default `~/go/bin`). That directory must be on your `PATH` to
+run them directly. If it isn't, add it:
+
+```bash
+gobin="$(go env GOBIN)"; [ -z "$gobin" ] && gobin="$(go env GOPATH)/bin"
+case ":$PATH:" in *":$gobin:"*) ;; *) export PATH="$PATH:$gobin" ;; esac
+```
+
 ### Cross-Compile Examples
 
 ```bash
@@ -396,6 +412,24 @@ go run ./cmd/zero-release build --goos windows --goarch amd64 --output dist/zero
 - [Performance](docs/PERFORMANCE.md)
 - [Agent evals](docs/AGENT_EVALS.md)
 
+## Community
+
+Questions, setup help, ideas, and sharing all live in
+[GitHub Discussions](https://github.com/Gitlawb/zero/discussions):
+
+| Category | Use it for |
+|---|---|
+| [Q&A](https://github.com/Gitlawb/zero/discussions/categories/q-a) | Setup help, provider/model configuration, "how do I" questions |
+| [Ideas](https://github.com/Gitlawb/zero/discussions/categories/ideas) | Feature proposals and design discussion before any PR |
+| [Show and tell](https://github.com/Gitlawb/zero/discussions/categories/show-and-tell) | Your skills, plugins, MCP setups, themes, and workflows |
+| [Announcements](https://github.com/Gitlawb/zero/discussions/categories/announcements) | Releases and project news from the maintainers |
+
+For a good Q&A answer fast, include `zero --version`, your OS and install
+method, the provider/model in use, and `zero doctor` output. See
+[SUPPORT.md](SUPPORT.md). Bugs belong in
+[issues](https://github.com/Gitlawb/zero/issues/new/choose); security reports
+follow [SECURITY.md](SECURITY.md), never a public thread.
+
 ## Contributing
 
 Contributions are welcome. Read [CONTRIBUTING.md](CONTRIBUTING.md), run the

SUPPORT.md

@@ -0,0 +1,36 @@
+# Getting help with Zero
+
+## Questions, setup help, configuration
+
+Ask in [GitHub Discussions](https://github.com/Gitlawb/zero/discussions). Use
+[Q&A](https://github.com/Gitlawb/zero/discussions/categories/q-a) so answers can
+be marked and found by the next person.
+
+The fastest way to a useful answer is to include:
+
+- `zero --version` output
+- your OS (Windows, macOS, Linux) and how you installed Zero (npm, install
+  script, source)
+- the provider and model you are using (`zero providers list`)
+- `zero doctor` output for setup problems
+- the exact command you ran and the exact error text
+
+## Bugs
+
+If Zero does something wrong (crashes, wrong output, broken behavior), file a
+[bug report](https://github.com/Gitlawb/zero/issues/new/choose). If you are not
+sure whether it is a bug or a setup problem, start in Discussions and we will
+promote it to an issue if it is one.
+
+## Feature ideas
+
+Start a thread in
+[Ideas](https://github.com/Gitlawb/zero/discussions/categories/ideas). Zero is
+in controlled development (see CONTRIBUTING.md), so discussing an idea first is
+also the path to getting a PR accepted.
+
+## Security issues
+
+Never file a public issue or discussion. Report privately via
+[GitHub Security Advisories](https://github.com/Gitlawb/zero/security/advisories/new)
+(see SECURITY.md).

bin/zero.js

@@ -55,6 +55,133 @@ function localControlHelperManifest(packageRoot) {
   return JSON.stringify({ version: 1, helpers });
 }
 
+// Mirrors internal/cli/observability.go parseDoctorArgs + writeDoctorHelp so the
+// wrapper's missing-binary doctor fallback matches the Go doctor CLI surface.
+const DOCTOR_HELP = `Usage:
+  zero doctor [flags]
+
+Runs Go backend health checks for config and provider setup.
+
+Flags:
+      --json            Print JSON report
+      --connectivity    Include provider endpoint connectivity probe when available
+  -h, --help            Show this help
+`;
+
+const EXIT_USAGE = 2;
+
+function installedByBun() {
+  if (process.env.ZERO_WRAPPER_SIMULATE_BUN === '1') {
+    return true;
+  }
+  return process.execPath.includes('bun') || !!process.versions?.bun;
+}
+
+function bunRecoveryParagraph() {
+  return (
+    'You installed with Bun, which does not run dependency lifecycle scripts\n' +
+    'by default. Trust the package to run the blocked postinstall:\n' +
+    '  bun pm trust @gitlawb/zero       (project install)\n' +
+    '  bun pm -g trust @gitlawb/zero    (global install)\n' +
+    'On Bun versions without `bun pm trust`, add\n' +
+    '  "trustedDependencies": ["@gitlawb/zero"]\n' +
+    'to your project package.json and reinstall.\n' +
+    '\n'
+  );
+}
+
+function buildFromSourceParagraph() {
+  return (
+    'If this platform has no prebuilt binary, build from source:\n' +
+    'https://github.com/Gitlawb/zero (go run ./cmd/zero, requires Go 1.26+).'
+  );
+}
+
+function missingBinaryContextParagraph() {
+  return (
+    'Normally npm installs it as an optional dependency of @gitlawb/zero\n' +
+    `(@gitlawb/zero-${process.platform}-${process.arch}), and the wrapper can\n` +
+    'also download it from the GitHub Release when it is missing.'
+  );
+}
+
+function missingNativeRecoveryParagraphs(postinstallScript) {
+  const ranByBun = installedByBun();
+  return (
+    missingBinaryContextParagraph() +
+    '\n' +
+    '\n' +
+    'Things to try:\n' +
+    '  - reinstall without omitting optional dependencies:\n' +
+    '      npm install -g @gitlawb/zero\n' +
+    '  - run the downloader manually (needs write access to the package dir):\n' +
+    `      node "${postinstallScript}"\n` +
+    '\n' +
+    (ranByBun ? bunRecoveryParagraph() : '') +
+    buildFromSourceParagraph()
+  );
+}
+
+function formatGenericMissingNativeBinaryMessage(postinstallScript) {
+  return (
+    '[zero] No native binary is available for this install.\n' +
+    missingNativeRecoveryParagraphs(postinstallScript)
+  );
+}
+
+function parseDoctorArgs(args) {
+  let json = false;
+  for (const arg of args) {
+    switch (arg) {
+      case '-h':
+      case '--help':
+      case 'help':
+        return { kind: 'help' };
+      case '--json':
+        json = true;
+        break;
+      case '--connectivity':
+        break;
+      default:
+        return { kind: 'error', message: `unknown doctor flag ${JSON.stringify(arg)}` };
+    }
+  }
+  return { kind: 'run', json };
+}
+
+function missingNativeDoctorJSONReport(postinstallScript) {
+  return {
+    generatedAt: new Date().toISOString().replace(/\.\d{3}Z$/, 'Z'),
+    ok: false,
+    checks: [
+      {
+        id: 'runtime.go',
+        label: 'Go runtime',
+        status: 'fail',
+        message: 'Native zero binary is missing next to the npm wrapper.',
+        details: {
+          remedy: `node "${postinstallScript}"`,
+        },
+      },
+    ],
+  };
+}
+
+function missingNativeDoctorTextReport(postinstallScript) {
+  return (
+    'Zero doctor report (' +
+    new Date().toISOString() +
+    ')\n' +
+    'Overall: fail\n' +
+    '[fail] runtime.go - Native zero binary is missing next to the npm wrapper.\n' +
+    '  remedy: node "' +
+    postinstallScript +
+    '"\n' +
+    '\n' +
+    missingNativeRecoveryParagraphs(postinstallScript)
+  );
+}
+
 // The platform payload is a version of @gitlawb/zero itself, installed under
 // an npm: alias (see docs/NPM_PACKAGING.md). The alias name is derived from
 // process.platform/process.arch, so an unsupported platform simply fails to
@@ -114,21 +241,38 @@ function resolveNativeBinary() {
 const nativePath = resolveNativeBinary();
 
 if (!nativePath) {
-  console.error(
-    '[zero] No native binary is available for this install.\n' +
-      'Normally npm installs it as an optional dependency of @gitlawb/zero\n' +
-      `(@gitlawb/zero-${process.platform}-${process.arch}), and the wrapper can\n` +
-      'also download it from the GitHub Release when it is missing.\n' +
-      '\n' +
-      'Things to try:\n' +
-      '  - reinstall without omitting optional dependencies:\n' +
-      '      npm install -g @gitlawb/zero\n' +
-      '  - run the downloader manually (needs write access to the package dir):\n' +
-      `      node "${join(packageRoot, 'scripts', 'postinstall.mjs')}"\n` +
-      '\n' +
-      'If this platform has no prebuilt binary, build from source:\n' +
-      'https://github.com/Gitlawb/zero (go run ./cmd/zero, requires Go 1.26+).',
-  );
+  const postinstallScript = join(packageRoot, 'scripts', 'postinstall.mjs');
+
+  // `zero doctor` is the diagnostic command: when the native binary is missing
+  // it's the one invocation that must NOT bail with the generic wrapper error,
+  // because that's exactly the blind alley issue #405 reports. Instead, surface
+  // a doctor-shaped FAIL line for the runtime so the user's diagnostic finds the
+  // real cause. We branch on a literal 'doctor' subcommand only (matching `zero
+  // doctor` and `zero doctor --connectivity`), preserving the existing bail for
+  // every other invocation (exec, providers list, TUI, --version, etc.).
+  const argv = process.argv.slice(2);
+  const isDoctor = argv.length > 0 && argv[0] === 'doctor';
+  if (isDoctor) {
+    const parsed = parseDoctorArgs(argv.slice(1));
+    if (parsed.kind === 'help') {
+      process.stdout.write(DOCTOR_HELP);
+      process.exit(0);
+    }
+    if (parsed.kind === 'error') {
+      process.stderr.write(`[zero] ${parsed.message}\n`);
+      process.exit(EXIT_USAGE);
+    }
+
+    if (parsed.json) {
+      process.stdout.write(JSON.stringify(missingNativeDoctorJSONReport(postinstallScript), null, 2) + '\n');
+      process.exit(1);
+    }
+
+    process.stdout.write(missingNativeDoctorTextReport(postinstallScript) + '\n');
+    process.exit(1);
+  }
+
+  console.error(formatGenericMissingNativeBinaryMessage(postinstallScript));
   process.exit(1);
 }
 

cmd/zero-perf-bench/main_test.go

@@ -4,6 +4,8 @@ import (
 	"bytes"
 	"strings"
 	"testing"
+
+	"github.com/Gitlawb/zero/internal/perfbench"
 )
 
 func TestParseArgsUsesCliAndEnvOverrides(t *testing.T) {
@@ -92,3 +94,18 @@ func TestHelpText(t *testing.T) {
 func emptyEnv(string) string {
 	return ""
 }
+
+func TestTurnExitCodeFailsWhenEveryTaskErrored(t *testing.T) {
+	var stderr bytes.Buffer
+	allErrored := perfbench.TurnBenchResult{TasksAttempted: 3, TasksErrored: 3}
+	if code := turnExitCode(allErrored, &stderr); code != 1 {
+		t.Fatalf("exit=%d, want 1 when every task errored", code)
+	}
+	if !strings.Contains(stderr.String(), "no accepted benchmark sample") {
+		t.Fatalf("stderr missing explanation: %q", stderr.String())
+	}
+	partial := perfbench.TurnBenchResult{TasksAttempted: 3, TasksErrored: 2}
+	if code := turnExitCode(partial, &stderr); code != 0 {
+		t.Fatalf("exit=%d, want 0 for partial errors (summary surfaces them)", code)
+	}
+}

cmd/zero-perf-bench/turn.go

@@ -9,6 +9,7 @@ import (
 	"path/filepath"
 	"strings"
 
+	"github.com/Gitlawb/zero/internal/execprofile"
 	"github.com/Gitlawb/zero/internal/perfbench"
 )
 
@@ -20,6 +21,7 @@ type turnOptions struct {
 	SuitePath   string
 	Model       string
 	Mode        string
+	ExecProfile string
 	SelfCorrect bool
 	Binary      string
 	Iterations  int
@@ -64,6 +66,7 @@ func runTurnCommand(args []string, getenv func(string) string, stdout io.Writer,
 	result, err := perfbench.RunTurnBench(context.Background(), set, perfbench.TurnBenchConfig{
 		Model:       options.Model,
 		Mode:        options.Mode,
+		ExecProfile: options.ExecProfile,
 		SelfCorrect: options.SelfCorrect,
 		Version:     options.Version,
 		Commit:      options.Commit,
@@ -86,9 +89,23 @@ func runTurnCommand(args []string, getenv func(string) string, stdout io.Writer,
 			_, _ = fmt.Fprintln(stderr, "[zero] Turn benchmark failed: "+err.Error())
 			return 1
 		}
-		return 0
+		return turnExitCode(result, stderr)
 	}
 	_, _ = fmt.Fprintln(stdout, perfbench.FormatTurnBenchSummary(result))
+	return turnExitCode(result, stderr)
+}
+
+// turnExitCode fails the command when every attempted task errored (no
+// iteration produced an accepted benchmark sample): such a report measures
+// nothing, and exiting 0 would let a broken configuration (missing binary, bad
+// path, failing harness step) pass as a clean baseline. Partial errors keep
+// exit 0 — the summary surfaces them loudly and the surviving samples are
+// still valid measurements.
+func turnExitCode(result perfbench.TurnBenchResult, stderr io.Writer) int {
+	if result.TasksAttempted > 0 && result.TasksErrored == result.TasksAttempted {
+		_, _ = fmt.Fprintln(stderr, "[zero] Turn benchmark failed: every task errored with no accepted benchmark sample (see warnings); the report contains no valid measurements")
+		return 1
+	}
 	return 0
 }
 
@@ -123,6 +140,13 @@ func parseTurnArgs(args []string, getenv func(string) string) (turnOptions, erro
 			}
 			options.Mode = value
 			index = next
+		case "--exec-profile":
+			value, next, err := readOptionValue(args, inlineValue, index, flag)
+			if err != nil {
+				return options, err
+			}
+			options.ExecProfile = value
+			index = next
 		case "--binary":
 			value, next, err := readOptionValue(args, inlineValue, index, flag)
 			if err != nil {
@@ -195,6 +219,20 @@ func parseTurnArgs(args []string, getenv func(string) string) (turnOptions, erro
 	if strings.TrimSpace(options.Model) == "" && !options.DryRun {
 		return options, fmt.Errorf("--model is required (or pass --dry-run)")
 	}
+	// Validate the profile here, before anything spawns: an unknown name would
+	// otherwise make every child exit with a usage error in milliseconds, and
+	// those near-zero walls would be recorded as valid latency samples in an
+	// exit-0 report — exactly the misread-as-improvement trap the harness
+	// closed for spawn failures. Normalizing to the catalog name also keeps
+	// the stamped execProfile canonical (Lookup is case-insensitive), so two
+	// captures of the same posture always compare equal.
+	if raw := strings.TrimSpace(options.ExecProfile); raw != "" {
+		profile, ok := execprofile.Lookup(raw)
+		if !ok {
+			return options, fmt.Errorf("unknown execution profile %q for --exec-profile. Valid profiles: %s", raw, strings.Join(execprofile.Names(), ", "))
+		}
+		options.ExecProfile = profile.Name
+	}
 	return options, nil
 }
 
@@ -225,6 +263,9 @@ func turnHelpText() string {
 		"  --suite <path>      Task set JSON file (required)",
 		"  --model <model>     Model to run (required unless --dry-run)",
 		"  --mode <name>       Exec mode preset to apply",
+		"  --exec-profile <name>",
+		"                      Execution profile for every task (fast|balanced|thorough);",
+		"                      forwarded to zero exec and stamped into the result",
 		"  --self-correct      Enable the post-edit verify-and-correct loop",
 		"  --binary <path>     Path to the `zero` binary (default: zero on PATH / repo root)",
 		"  --iterations <n>    Times to run each task (default: 1)",

cmd/zero-perf-bench/turn_test.go

@@ -0,0 +1,65 @@
+package main
+
+import (
+	"strings"
+	"testing"
+)
+
+func TestParseTurnArgsExecProfile(t *testing.T) {
+	getenv := func(string) string { return "" }
+	options, err := parseTurnArgs([]string{
+		"--suite", "suite.json",
+		"--model", "m",
+		"--exec-profile", "fast",
+	}, getenv)
+	if err != nil {
+		t.Fatalf("parseTurnArgs: %v", err)
+	}
+	if options.ExecProfile != "fast" {
+		t.Fatalf("ExecProfile = %q, want fast", options.ExecProfile)
+	}
+
+	// Inline form.
+	options, err = parseTurnArgs([]string{
+		"--suite=suite.json",
+		"--model=m",
+		"--exec-profile=thorough",
+	}, getenv)
+	if err != nil {
+		t.Fatalf("parseTurnArgs inline: %v", err)
+	}
+	if options.ExecProfile != "thorough" {
+		t.Fatalf("ExecProfile = %q, want thorough", options.ExecProfile)
+	}
+}
+
+// The bench validates the profile BEFORE spawning anything: an unknown name
+// would otherwise make every child exit with a usage error in milliseconds
+// and record those near-zero walls as valid latency samples in an exit-0
+// report. Lookup normalization also keeps the stamped name canonical so two
+// captures of the same posture compare equal.
+func TestParseTurnArgsExecProfileValidatesAndNormalizes(t *testing.T) {
+	getenv := func(string) string { return "" }
+
+	options, err := parseTurnArgs([]string{
+		"--suite", "suite.json", "--model", "m", "--exec-profile", "FAST",
+	}, getenv)
+	if err != nil {
+		t.Fatalf("parseTurnArgs: %v", err)
+	}
+	if options.ExecProfile != "fast" {
+		t.Fatalf("ExecProfile = %q, want the canonical fast", options.ExecProfile)
+	}
+
+	_, err = parseTurnArgs([]string{
+		"--suite", "suite.json", "--model", "m", "--exec-profile", "blanced",
+	}, getenv)
+	if err == nil {
+		t.Fatal("an unknown profile must fail parse, before anything spawns")
+	}
+	for _, name := range []string{"balanced", "fast", "thorough"} {
+		if !strings.Contains(err.Error(), name) {
+			t.Fatalf("error must list %q, got %q", name, err.Error())
+		}
+	}
+}

internal/agent/compaction.go

@@ -56,6 +56,10 @@ type CompactionOptions struct {
 	// injected so Compact stays pure and testable; the agent loop wires it to a
 	// real provider call.
 	Summarize func(toSummarize []zeroruntime.Message) (string, error)
+	// taskState is a snapshot supplied by the running agent. Its immutable
+	// objective is always preserved; mutable fields are admitted only when its
+	// plan projection still matches the transcript.
+	taskState *taskStateSnapshot
 }
 
 // CompactionResult is the metadata-bearing result returned by CompactMessages.
@@ -220,7 +224,7 @@ func CompactMessages(messages []zeroruntime.Message, opts CompactionOptions) (Co
 
 	// Preserve structured state (active plan + loaded skills) from the elided
 	// middle verbatim, so it is not lost or paraphrased away by the prose summary.
-	content := appendPreservedState(summaryLabel+"\n"+summary, middle)
+	content := appendPreservedState(summaryLabel+"\n"+summary, middle, opts.taskState)
 
 	compacted := make([]zeroruntime.Message, 0, systemEnd+1+(len(messages)-boundary))
 	compacted = append(compacted, messages[:systemEnd]...)
@@ -326,6 +330,7 @@ type compactionState struct {
 	// OnText is deliberately NOT forwarded (compaction stays invisible to the user),
 	// but its token COST must still be counted so usage reports and budgets include it.
 	onUsage func(Usage)
+	task    *taskState
 
 	// calibrationRatio scales the raw byte/4 token estimate toward the provider's
 	// real prompt-token count. ApproxTextTokens over-counts code-heavy content by
@@ -365,13 +370,15 @@ func (state *compactionState) calibratedTokens(raw int) int {
 	return int(float64(raw) * state.calibrationRatio)
 }
 
-func newCompactionState(options Options) *compactionState {
-	return &compactionState{
+func newCompactionState(options Options, task *taskState) *compactionState {
+	state := &compactionState{
 		enabled:      options.ContextWindow > 0,
 		threshold:    compactionThreshold(options.ContextWindow),
 		preserveLast: options.CompactionPreserveLast,
 		onUsage:      options.OnUsage,
+		task:         task,
 	}
+	return state
 }
 
 // maybeCompact runs proactive compaction at the top of a turn. It returns the
@@ -418,6 +425,7 @@ func (state *compactionState) maybeCompact(
 	compacted, err := Compact(messages, CompactionOptions{
 		PreserveLast: state.preserveLast,
 		Summarize:    summarizeClosure(ctx, provider, state.onUsage),
+		taskState:    state.task.snapshotForCompaction(messages),
 	})
 	if err != nil {
 		// Summarizer failed: keep the original history. The reactive path (or a
@@ -468,6 +476,7 @@ func (state *compactionState) recover(
 	result, compactErr := Compact(messages, CompactionOptions{
 		PreserveLast: state.preserveLast,
 		Summarize:    summarizeClosure(ctx, provider, state.onUsage),
+		taskState:    state.task.snapshotForCompaction(messages),
 	})
 	if compactErr != nil {
 		// A genuine compaction attempt was made (and failed): the budget is spent

internal/agent/compaction_preserve.go

@@ -30,6 +30,10 @@ const (
 const (
 	maxRecentEdits   = 20
 	maxEditNoteBytes = 160
+	// maxTaskObjectiveBytes keeps the original objective recognizable after
+	// compaction without allowing an unusually large prompt to recreate the
+	// context pressure the compaction just removed.
+	maxTaskObjectiveBytes = 512
 )
 
 const (
@@ -76,32 +80,14 @@ func extractLatestPlan(messages []zeroruntime.Message) string {
 // ({"plan":[{content,status,...}]}) as terse status-tagged bullet lines. Returns
 // "" on malformed arguments or an empty plan.
 func formatPlanArguments(arguments string) string {
-	var parsed struct {
-		Plan []struct {
-			Content string `json:"content"`
-			Step    string `json:"step"`
-			Status  string `json:"status"`
-			Notes   string `json:"notes"`
-		} `json:"plan"`
-	}
-	if err := json.Unmarshal([]byte(strings.TrimSpace(arguments)), &parsed); err != nil {
+	plan, ok := parseTaskPlan(arguments)
+	if !ok {
 		return ""
 	}
-	lines := make([]string, 0, len(parsed.Plan))
-	for _, item := range parsed.Plan {
-		content := strings.TrimSpace(item.Content)
-		if content == "" {
-			content = strings.TrimSpace(item.Step)
-		}
-		if content == "" {
-			continue
-		}
-		status := strings.TrimSpace(item.Status)
-		if status == "" {
-			status = "pending"
-		}
-		line := "- [" + status + "] " + content
-		if notes := strings.TrimSpace(item.Notes); notes != "" {
+	lines := make([]string, 0, len(plan))
+	for _, item := range plan {
+		line := "- [" + item.Status + "] " + item.Content
+		if notes := item.Notes; notes != "" {
 			line += "\n  Notes: " + notes
 		}
 		lines = append(lines, line)
@@ -431,12 +417,25 @@ func capBody(body string) string {
 // preservedState is the JSON shape of the carried-across-compaction block.
 type preservedState struct {
 	Plan                string                 `json:"plan,omitempty"`
+	Task                *preservedTaskState    `json:"task,omitempty"`
 	RecentEdits         []preservedEdit        `json:"recent_edits,omitempty"`
 	Tools               []preservedTool        `json:"tools,omitempty"`
 	Skills              []preservedSkill       `json:"skills,omitempty"`
 	ProjectInstructions []preservedInstruction `json:"project_instructions,omitempty"`
 }
 
+type preservedTaskState struct {
+	Objective           string     `json:"objective"`
+	Status              taskStatus `json:"status,omitempty"`
+	Pending             int        `json:"pending,omitempty"`
+	InProgress          int        `json:"in_progress,omitempty"`
+	Completed           int        `json:"completed,omitempty"`
+	Failed              int        `json:"failed,omitempty"`
+	VerificationPassed  int        `json:"verification_passed,omitempty"`
+	VerificationFailed  int        `json:"verification_failed,omitempty"`
+	VerificationOutcome Outcome    `json:"verification_outcome,omitempty"`
+}
+
 type preservedEdit struct {
 	Path string `json:"path"`
 	Note string `json:"note,omitempty"`
@@ -464,8 +463,27 @@ type preservedInstruction struct {
 // may live only inside the injected summary message, which on a later compaction
 // lands in middle with no real tool calls left to extract. Fresh tool calls and
 // instruction blocks override the carried-forward copy by name/source.
-func appendPreservedState(summary string, middle []zeroruntime.Message) string {
+func appendPreservedState(summary string, middle []zeroruntime.Message, taskSnapshot *taskStateSnapshot) string {
 	priorState := parsePreservedStateBlock(latestSummaryContent(middle))
+	task := priorState.Task
+	if taskSnapshot != nil {
+		task = &preservedTaskState{
+			Objective: capTaskObjective(taskSnapshot.Objective),
+		}
+		// Plan parity corroborates only the mutable task projection. The objective
+		// comes directly from the run prompt and is immutable, so it must survive
+		// even after compaction removes the plan tool call needed for comparison.
+		if taskSnapshot.PlanParity == taskPlanParityMatch {
+			task.Status = taskSnapshot.Status
+			task.Pending = taskSnapshot.Plan.Pending
+			task.InProgress = taskSnapshot.Plan.InProgress
+			task.Completed = taskSnapshot.Plan.Completed
+			task.Failed = taskSnapshot.Plan.Failed
+			task.VerificationPassed = taskSnapshot.Verification.Passed
+			task.VerificationFailed = taskSnapshot.Verification.Failed
+			task.VerificationOutcome = taskSnapshot.Verification.LastOutcome
+		}
+	}
 
 	// Plan: a fresh update_plan in middle is authoritative; otherwise carry the
 	// plan preserved by an earlier compaction.
@@ -493,12 +511,25 @@ func appendPreservedState(summary string, middle []zeroruntime.Message) string {
 		projectInstructionEntries(middle),
 	)
 
-	if block := formatPreservedState(plan, edits, tools, skills, instructions); block != "" {
+	if block := formatPreservedState(plan, task, edits, tools, skills, instructions); block != "" {
 		summary += "\n\n" + block
 	}
 	return summary
 }
 
+func capTaskObjective(objective string) string {
+	objective = strings.TrimSpace(objective)
+	if len(objective) <= maxTaskObjectiveBytes {
+		return objective
+	}
+	const suffix = "…"
+	limit := maxTaskObjectiveBytes - len(suffix)
+	for limit > 0 && !utf8.RuneStart(objective[limit]) {
+		limit--
+	}
+	return strings.TrimSpace(objective[:limit]) + suffix
+}
+
 // mergeRecentEdits overlays fresh edits onto edits preserved by an earlier
 // compaction. Unlike mergeSkillEntries (which keeps refreshed entries in their
 // original slot), a path touched again by a fresh edit MOVES to the newest
@@ -555,11 +586,11 @@ func preservedEditsToEntries(edits []preservedEdit) []skillEntry {
 
 // formatPreservedState renders state as the labelled, single-line
 // JSON block. Returns "" when there is nothing to preserve.
-func formatPreservedState(plan string, edits, tools, skills, instructions []skillEntry) string {
-	if plan == "" && len(edits) == 0 && len(tools) == 0 && len(skills) == 0 && len(instructions) == 0 {
+func formatPreservedState(plan string, task *preservedTaskState, edits, tools, skills, instructions []skillEntry) string {
+	if plan == "" && task == nil && len(edits) == 0 && len(tools) == 0 && len(skills) == 0 && len(instructions) == 0 {
 		return ""
 	}
-	state := preservedState{Plan: plan}
+	state := preservedState{Plan: plan, Task: task}
 	for _, e := range edits {
 		state.RecentEdits = append(state.RecentEdits, preservedEdit{Path: e.name, Note: e.body})
 	}

internal/agent/compaction_preserve_test.go

@@ -1,6 +1,7 @@
 package agent
 
 import (
+	"encoding/json"
 	"strings"
 	"testing"
 	"unicode/utf8"
@@ -59,6 +60,103 @@ func TestCompactPreservesActivePlan(t *testing.T) {
 	}
 }
 
+func TestCompactPreservesBoundedTaskContext(t *testing.T) {
+	objective := strings.Repeat("世", maxTaskObjectiveBytes)
+	task := newTaskState(objective, nil)
+	task.observe(taskStateEvent{kind: taskStateEventPlan, arguments: `{"plan":[{"content":"write code","status":"in_progress"},{"content":"add tests","status":"pending"}]}`})
+	messages := stateConversation()
+	compacted, err := Compact(messages, CompactionOptions{
+		PreserveLast: 2,
+		Summarize:    func([]zeroruntime.Message) (string, error) { return "SUMMARY", nil },
+		taskState:    task.snapshotForCompaction(messages),
+	})
+	if err != nil {
+		t.Fatalf("Compact: %v", err)
+	}
+	state := parsePreservedStateBlock(compacted[1].Content)
+	if state.Task == nil || state.Task.Status != taskStatusActive || state.Task.InProgress != 1 || state.Task.Pending != 1 {
+		t.Fatalf("unexpected compact task state: %#v", state.Task)
+	}
+	if len(state.Task.Objective) > maxTaskObjectiveBytes || !utf8.ValidString(state.Task.Objective) {
+		t.Fatalf("objective was not safely bounded: %d bytes %q", len(state.Task.Objective), state.Task.Objective)
+	}
+}
+
+func TestCompactPreservesObjectiveAfterPlanParityMismatch(t *testing.T) {
+	prior := preservedState{Task: &preservedTaskState{Objective: "stale objective", Status: taskStatusActive, Pending: 1}}
+	encoded, err := json.Marshal(prior)
+	if err != nil {
+		t.Fatal(err)
+	}
+	messages := []zeroruntime.Message{
+		{Role: zeroruntime.MessageRoleSystem, Content: "system"},
+		{Role: zeroruntime.MessageRoleUser, Content: summaryLabel + "\nold\n\n" + preservedStateLabel + "\n" + string(encoded)},
+		{Role: zeroruntime.MessageRoleAssistant, Content: "continuing"},
+		{Role: zeroruntime.MessageRoleUser, Content: "more"},
+		{Role: zeroruntime.MessageRoleAssistant, Content: "working"},
+		{Role: zeroruntime.MessageRoleUser, Content: "again"},
+		{Role: zeroruntime.MessageRoleAssistant, Content: "done"},
+	}
+	compacted, err := Compact(messages, CompactionOptions{
+		PreserveLast: 2,
+		Summarize:    func([]zeroruntime.Message) (string, error) { return "SUMMARY", nil },
+		taskState: &taskStateSnapshot{
+			Objective:  "current objective",
+			PlanParity: taskPlanParityMismatch,
+		},
+	})
+	if err != nil {
+		t.Fatalf("Compact: %v", err)
+	}
+	state := parsePreservedStateBlock(compacted[1].Content)
+	if state.Task == nil || state.Task.Objective != "current objective" {
+		t.Fatalf("immutable objective was lost on plan mismatch: %#v", state.Task)
+	}
+	if state.Task.Status != "" || state.Task.Pending != 0 {
+		t.Fatalf("uncorroborated mutable task fields survived plan mismatch: %#v", state.Task)
+	}
+}
+
+func TestTaskObjectiveSurvivesRepeatedCompactionWithoutPlanRefresh(t *testing.T) {
+	task := newTaskState("keep this objective", nil)
+	task.observe(taskStateEvent{kind: taskStateEventPlan, arguments: `{"plan":[{"content":"write code","status":"in_progress"},{"content":"add tests","status":"pending"}]}`})
+	messages := stateConversation()
+
+	first, err := Compact(messages, CompactionOptions{
+		PreserveLast: 2,
+		Summarize:    func([]zeroruntime.Message) (string, error) { return "FIRST", nil },
+		taskState:    task.snapshotForCompaction(messages),
+	})
+	if err != nil {
+		t.Fatalf("first Compact: %v", err)
+	}
+	secondInput := append(append([]zeroruntime.Message{}, first...),
+		zeroruntime.Message{Role: zeroruntime.MessageRoleUser, Content: "more"},
+		zeroruntime.Message{Role: zeroruntime.MessageRoleAssistant, Content: "working"},
+		zeroruntime.Message{Role: zeroruntime.MessageRoleUser, Content: "again"},
+		zeroruntime.Message{Role: zeroruntime.MessageRoleAssistant, Content: "done"},
+	)
+	snapshot := task.snapshotForCompaction(secondInput)
+	if snapshot.PlanParity != taskPlanParityMismatch {
+		t.Fatalf("plan call should be absent after first compaction, parity=%q", snapshot.PlanParity)
+	}
+	second, err := Compact(secondInput, CompactionOptions{
+		PreserveLast: 2,
+		Summarize:    func([]zeroruntime.Message) (string, error) { return "SECOND", nil },
+		taskState:    snapshot,
+	})
+	if err != nil {
+		t.Fatalf("second Compact: %v", err)
+	}
+	state := parsePreservedStateBlock(second[1].Content)
+	if state.Task == nil || state.Task.Objective != "keep this objective" {
+		t.Fatalf("objective lost after second compaction: %#v", state.Task)
+	}
+	if state.Task.Status != "" || state.Task.InProgress != 0 {
+		t.Fatalf("uncorroborated mutable fields survived second compaction: %#v", state.Task)
+	}
+}
+
 func TestCompactPreservesLoadedSkills(t *testing.T) {
 	summary := compactStateConversation(t, stateConversation())
 	if !strings.Contains(summary, preservedStateLabel) {
@@ -283,12 +381,10 @@ func TestExtractLatestPlanReturnsMostRecent(t *testing.T) {
 	}
 }
 
-func TestFormatPlanArgumentsAcceptsStepAlias(t *testing.T) {
+func TestFormatPlanArgumentsRejectsUnsupportedStepAlias(t *testing.T) {
 	got := formatPlanArguments(`{"plan":[{"step":"write failing test","status":"in_progress"},{"content":"keep existing shape","status":"pending"}]}`)
-	for _, want := range []string{"- [in_progress] write failing test", "- [pending] keep existing shape"} {
-		if !strings.Contains(got, want) {
-			t.Fatalf("expected %q in formatted plan, got %q", want, got)
-		}
+	if got != "" {
+		t.Fatalf("unsupported alias should reject the whole plan like update_plan, got %q", got)
 	}
 }
 

internal/agent/compaction_test.go

@@ -581,7 +581,7 @@ func TestCompactNeverProducesConsecutiveUserMessages(t *testing.T) {
 }
 
 func TestRecoverNoopDoesNotConsumeReactiveBudget(t *testing.T) {
-	st := newCompactionState(Options{ContextWindow: 1000, CompactionPreserveLast: 2})
+	st := newCompactionState(Options{ContextWindow: 1000, CompactionPreserveLast: 2}, nil)
 	provider := &mockProvider{turns: [][]zeroruntime.StreamEvent{{
 		{Type: zeroruntime.StreamEventText, Content: "SUMMARY"}, {Type: zeroruntime.StreamEventDone},
 	}}}
@@ -628,7 +628,7 @@ func TestRecoverNoopDoesNotConsumeReactiveBudget(t *testing.T) {
 }
 
 func TestRecoverDisabledIsNoop(t *testing.T) {
-	st := newCompactionState(Options{ContextWindow: 0})
+	st := newCompactionState(Options{ContextWindow: 0}, nil)
 	msgs := []zeroruntime.Message{{Role: zeroruntime.MessageRoleUser, Content: "x"}}
 	called := false
 	// recover must not invoke the provider/summarize when disabled, even on a

internal/agent/completion_policy.go

@@ -39,7 +39,7 @@ func newCompletionPolicy(requireSemanticCheck bool) *completionPolicy {
 	return &completionPolicy{requireSemanticCheck: requireSemanticCheck}
 }
 
-func (policy *completionPolicy) evaluate(text string, planPending bool) completionEvaluation {
+func (policy *completionPolicy) evaluate(text string, context completionContext) completionEvaluation {
 	// A direct admission is strong local evidence and takes precedence over all
 	// weaker signals, avoiding both continuation nudges and semantic checks.
 	if reason := selfReportedIncompletion(text); reason != "" {
@@ -50,7 +50,7 @@ func (policy *completionPolicy) evaluate(text string, planPending bool) completi
 	// weak evidence because bookkeeping can be stale. Both get a bounded chance
 	// to continue, but only a persisted cue is ultimately incomplete.
 	cue := endsWithContinuationCue(text)
-	if cue || planPending {
+	if cue || context.PlanPending {
 		if policy.continueNudges < maxContinueNudges {
 			policy.continueNudges++
 			reason := "your message ended mid-step"

internal/agent/completion_policy_test.go

@@ -1,16 +1,19 @@
 package agent
 
-import "testing"
+import (
+	"strings"
+	"testing"
+)
 
 func TestCompletionPolicyLocalEvidenceDecidesWithoutSemanticCheck(t *testing.T) {
 	policy := newCompletionPolicy(false)
 
-	complete := policy.evaluate("Done. All required checks pass.", false)
+	complete := policy.evaluate("Done. All required checks pass.", completionContext{})
 	if complete.Decision != CompletionComplete {
 		t.Fatalf("confident completion decision = %q, want %q", complete.Decision, CompletionComplete)
 	}
 
-	incomplete := policy.evaluate("I couldn't verify the result, so this is my best guess.", false)
+	incomplete := policy.evaluate("I couldn't verify the result, so this is my best guess.", completionContext{})
 	if incomplete.Decision != CompletionIncomplete {
 		t.Fatalf("admitted failure decision = %q, want %q", incomplete.Decision, CompletionIncomplete)
 	}
@@ -19,13 +22,13 @@ func TestCompletionPolicyLocalEvidenceDecidesWithoutSemanticCheck(t *testing.T)
 func TestCompletionPolicyPreservesBoundedPlanStallProtection(t *testing.T) {
 	policy := newCompletionPolicy(false)
 	for attempt := 0; attempt < maxContinueNudges; attempt++ {
-		got := policy.evaluate("Let me inspect the remaining configuration:", true)
+		got := policy.evaluate("Let me inspect the remaining configuration:", completionContext{PlanPending: true})
 		if got.Decision != CompletionUncertain || got.Action != completionActionContinue {
 			t.Fatalf("attempt %d = (%q, %q), want uncertain continue", attempt+1, got.Decision, got.Action)
 		}
 	}
 
-	got := policy.evaluate("Let me inspect the remaining configuration:", true)
+	got := policy.evaluate("Let me inspect the remaining configuration:", completionContext{PlanPending: true})
 	if got.Decision != CompletionIncomplete {
 		t.Fatalf("decision after nudge budget = %q, want %q", got.Decision, CompletionIncomplete)
 	}
@@ -34,13 +37,13 @@ func TestCompletionPolicyPreservesBoundedPlanStallProtection(t *testing.T) {
 func TestCompletionPolicyTreatsPendingPlanAsWeakEvidence(t *testing.T) {
 	policy := newCompletionPolicy(false)
 	for attempt := 0; attempt < maxContinueNudges; attempt++ {
-		got := policy.evaluate("All set.", true)
+		got := policy.evaluate("All set.", completionContext{PlanPending: true})
 		if got.Decision != CompletionUncertain || got.Action != completionActionContinue {
 			t.Fatalf("attempt %d = (%q, %q), want uncertain continue", attempt+1, got.Decision, got.Action)
 		}
 	}
 
-	got := policy.evaluate("All set.", true)
+	got := policy.evaluate("All set.", completionContext{PlanPending: true})
 	if got.Decision != CompletionComplete {
 		t.Fatalf("stale plan decision after nudge budget = %q, want %q", got.Decision, CompletionComplete)
 	}
@@ -49,12 +52,12 @@ func TestCompletionPolicyTreatsPendingPlanAsWeakEvidence(t *testing.T) {
 func TestCompletionPolicyAllowsExactlyOneRequiredSemanticCheck(t *testing.T) {
 	policy := newCompletionPolicy(true)
 
-	first := policy.evaluate("Implemented and tested.", false)
+	first := policy.evaluate("Implemented and tested.", completionContext{})
 	if first.Decision != CompletionUncertain || first.Action != completionActionSemanticCheck {
 		t.Fatalf("first decision = (%q, %q), want uncertain semantic check", first.Decision, first.Action)
 	}
 
-	second := policy.evaluate("PASS. The result meets the task criterion.", false)
+	second := policy.evaluate("PASS. The result meets the task criterion.", completionContext{})
 	if second.Decision != CompletionComplete {
 		t.Fatalf("post-check decision = %q, want %q", second.Decision, CompletionComplete)
 	}
@@ -62,3 +65,11 @@ func TestCompletionPolicyAllowsExactlyOneRequiredSemanticCheck(t *testing.T) {
 		t.Fatal("semantic check was requested more than once")
 	}
 }
+
+func TestAcceptanceVerificationNudgeIncludesBoundedObjective(t *testing.T) {
+	objective := "make the requested behavior observable"
+	nudge := acceptanceVerificationNudge(objective)
+	if !strings.Contains(nudge, "Task objective: "+objective) {
+		t.Fatalf("semantic check was not grounded in the objective: %q", nudge)
+	}
+}

internal/agent/guardrails.go

@@ -1,7 +1,6 @@
 package agent
 
 import (
-	"encoding/json"
 	"strconv"
 	"strings"
 
@@ -127,9 +126,8 @@ func endsWithContinuationCue(text string) bool {
 // which the update_plan tool coerces to "pending") counts as remaining, matching
 // that tool's own status normalization.
 func planStatusRemaining(status string) bool {
-	switch strings.ToLower(strings.TrimSpace(status)) {
-	case "completed", "complete", "done", "finished", "resolved", "✓", "x", "[x]",
-		"failed", "fail", "error", "errored", "blocked", "cancelled", "canceled", "abandoned", "skipped":
+	switch normalizeTaskPlanStatus(status) {
+	case "completed", "failed":
 		return false
 	default:
 		return true
@@ -307,8 +305,8 @@ func hasAnyPrefix(s string, prefixes []string) bool {
 // the bug-hunt surfaced: well-formed output treated as correct; pre-existing tests
 // passing treated as the objective being met; and a result that merely matches a
 // baseline the task asked to beat or improve. General — no task-specific content.
-func acceptanceVerificationNudge() string {
-	return "Before this task can be marked complete, " + acceptanceNudgeMarker + " — " +
+func acceptanceVerificationNudge(objective string) string {
+	nudge := "Before this task can be marked complete, " + acceptanceNudgeMarker + " — " +
 		"NOT the shape or format of your output, NOT that pre-existing tests pass, and NOT that your " +
 		"result merely matches a baseline you were asked to beat or improve. " +
 		"Re-read the original task, then run a concrete check that exercises the actual requirement: " +
@@ -316,6 +314,10 @@ func acceptanceVerificationNudge() string {
 		"thing the task asked you to produce, recover, fix, or optimize. " +
 		"If that check passes, reply PASS and cite the evidence. " +
 		"If it does not pass — or you cannot run such a check — say so plainly and keep working; do not claim success."
+	if objective = capTaskObjective(objective); objective != "" {
+		nudge += "\n\nTask objective: " + objective
+	}
+	return nudge
 }
 
 // toolFailureHintMarker is a stable substring for tests.
@@ -553,16 +555,12 @@ func (state *guardState) pendingPlanItems() bool {
 // many items are still remaining. Malformed arguments leave the prior count
 // unchanged (best-effort — the plan panel itself tolerates the same).
 func (state *guardState) observePlanUpdate(arguments string) {
-	var parsed struct {
-		Plan []struct {
-			Status string `json:"status"`
-		} `json:"plan"`
-	}
-	if json.Unmarshal([]byte(arguments), &parsed) != nil {
+	plan, ok := parseTaskPlan(arguments)
+	if !ok {
 		return
 	}
 	pending := 0
-	for _, item := range parsed.Plan {
+	for _, item := range plan {
 		if planStatusRemaining(item.Status) {
 			pending++
 		}

internal/agent/guardrails_test.go

@@ -169,6 +169,37 @@ func TestRunResetsEmptyTurnCounterOnToolCall(t *testing.T) {
 	}
 }
 
+// The whole point of #702's fix: the unknown-session error's normalized
+// signature must not vary with the session id, so a model probing ids 1, 2,
+// 3, … produces one repeated signature the failure guard can count. If the id
+// leaked into the first 80 normalized chars, each probe would reset the streak
+// and the halt would never fire.
+func TestUnknownExecSessionErrorSignatureIsIDInvariant(t *testing.T) {
+	a := errorSignature(tools.UnknownExecSessionError(1))
+	b := errorSignature(tools.UnknownExecSessionError(999999))
+	if a != b {
+		t.Fatalf("unknown-session signature varies with id:\n  %q\n  %q", a, b)
+	}
+}
+
+// End to end: with an id-invariant signature, probing a different unknown id
+// each turn now trips the repeated-failure halt at toolFailureStopAt, where
+// before the fix it never would.
+func TestUnknownExecSessionProbingTripsFailureHalt(t *testing.T) {
+	var state guardState
+	var stoppedAt int
+	for i := 1; i <= toolFailureStopAt; i++ {
+		out := state.observeToolResult(tools.WriteStdinToolName, true, tools.UnknownExecSessionError(i))
+		if out.Stop {
+			stoppedAt = i
+			break
+		}
+	}
+	if stoppedAt != toolFailureStopAt {
+		t.Fatalf("probing distinct unknown ids stopped at %d, want %d", stoppedAt, toolFailureStopAt)
+	}
+}
+
 func TestGuardStateResetsToolOnlyStreakOnEmptyNonToolTurn(t *testing.T) {
 	var state guardState
 	toolOnly := zeroruntime.CollectedStream{

internal/agent/loop.go

@@ -136,6 +136,33 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 		}
 	}
 
+	// Establish the run's turn session — the seam an optimized provider session
+	// (connection reuse, prewarm, native compaction) plugs into without touching
+	// this loop. Default: wrap the passed provider so the session's Stream IS
+	// provider.StreamCompletion, byte-identical to calling it directly. Opened
+	// before dispatchSessionStart so a failed open is a clean run-start error
+	// with no session hooks fired and nothing to unwind.
+	turnSessions := options.TurnSessionProvider
+	if turnSessions == nil {
+		turnSessions = zeroruntime.NewProviderTurnSessionProvider(provider, zeroruntime.ProviderCapabilities{})
+	}
+	session, sessionErr := turnSessions.OpenTurnSession(ctx)
+	if sessionErr != nil {
+		return Result{}, fmt.Errorf("agent: open turn session: %w", sessionErr)
+	}
+	// Closed via closure (not `defer session.Close()`) so after a mid-run model
+	// swap the CURRENT session is the one closed at teardown — the swap closes
+	// the old one inline. Close errors are advisory and never fail the run.
+	defer func() { _ = session.Close() }()
+	// Best-effort prewarm: the default adapter no-ops; a real session may prime
+	// a connection. Failure is ignored by contract (never fatal).
+	_ = session.Prewarm(ctx)
+	// Route ALL provider I/O — per-turn streams, mid-stream retries, the
+	// compaction summarizer, and the final-answer request — through the session
+	// by reassigning the loop's provider local. Every existing call site reads
+	// this local, so no signatures change anywhere downstream.
+	provider = sessionProvider{session: session}
+
 	maxTurns := options.MaxTurns
 	if maxTurns <= 0 {
 		maxTurns = 12
@@ -154,10 +181,46 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 	options.runPermissions = runPermissions
 	defer runPermissions.cleanup()
 
-	messages := zeroruntime.SeedMessagesWithImages(buildSystemPrompt(options), prompt, options.Images)
+	promptParts := buildSystemPromptParts(options)
+	messages := zeroruntime.SeedMessagesWithImages(promptParts.prompt, prompt, options.Images)
 
 	guards := newGuardState()
-	compactor := newCompactionState(options)
+	task := newTaskState(prompt, options.Trace)
+	compactor := newCompactionState(options, task)
+	defer func() {
+		// A final transcript comparison is observational. It records drift but
+		// never rewrites the result or changes execution after the fact.
+		task.observePlanParity(result.Messages)
+		// Tool-result observations are intentionally coalesced; this final event
+		// guarantees their aggregate counts are present even on an early return.
+		task.emit()
+	}()
+	// The execution-profile posture controller. nil Options.Profile (every
+	// existing caller) makes every observe/decide call a no-op, keeping the
+	// loop byte-identical for unprofiled runs.
+	posture := newProfileController(options.Profile)
+	// applyPosture applies at most one posture escalation when an armed trigger
+	// has fired: raise the hoisted turn ceiling, restore effort and the
+	// completion gate, stamp the counter. Shared by the end-of-turn tail and
+	// the completion-gate uncertain path (which continues the loop before the
+	// tail runs). No-op forever after the first escalation and for unprofiled
+	// runs.
+	applyPosture := func() {
+		if target, fired := posture.maybeEscalate(); fired {
+			if target.MaxTurns > maxTurns {
+				maxTurns = target.MaxTurns
+			}
+			if target.ReasoningEffort != "" {
+				options.ReasoningEffort = target.ReasoningEffort
+			} else if target.RestoreDefaultEffort {
+				options.ReasoningEffort = ""
+			}
+			if target.RestoreCompletionGate {
+				options.RequireCompletionSignal = true
+			}
+			options.Trace.Counter(trace.CounterPostureEscalations, 1)
+		}
+	}
 
 	// Background post-edit diagnostics: files changed by mutating tools are
 	// checked off the tool-call critical path and any errors are appended as a
@@ -215,16 +278,14 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 		exposed, _ := partitionToolsCached(registry, permissionMode, options, loaded, toolDefCache)
 		toolPartitionSpan.End()
 
-		// Prompt-prefix fingerprint: hash the seven cacheable sub-components
-		// of this turn's request (system prompt sections, project context,
-		// skills, partitioned tool list, tool schemas) and emit one trace
-		// event. Stable across turns => cacheable. Drift in any sub-hash
-		// names the sub-component that broke the cache for this turn. The
-		// computation is pure and cheap (seven SHA-256s of small strings)
-		// and is a no-op when tracing is off.
+		// Fingerprint the exact system prompt built at run start plus this turn's
+		// provider-visible tool definitions in their emitted order. Reusing the
+		// retained prompt parts avoids rebuilding workspace context while ensuring
+		// the trace describes the same bytes carried by the request.
 		if options.Trace != nil {
-			fp := ComputePrefixFingerprint(options, exposed)
+			fp := computePrefixFingerprint(buildPromptSubstringsFromParts(promptParts, exposed))
 			options.Trace.EmitPrefixHash(trace.PrefixHash{
+				SystemPromptHash:       fp.SystemPromptHash,
 				BaseInstructionsHash:   fp.BaseInstructionsHash,
 				ConfirmationPolicyHash: fp.ConfirmationPolicyHash,
 				ProjectContextHash:     fp.ProjectContextHash,
@@ -518,7 +579,9 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 			// model's final answer ONLY when the work is actually done. Default off
 			// (RequireCompletionSignal), so interactive runs stay byte-identical.
 			if options.RequireCompletionSignal {
-				evaluation := completionPolicy.evaluate(collected.Text, guards.pendingPlanItems())
+				completionContext := task.completionContext(messages, guards.pendingPlanItems())
+				evaluation := completionPolicy.evaluate(collected.Text, completionContext)
+				task.observe(taskStateEvent{kind: taskStateEventCompletion, completion: evaluation})
 				switch evaluation.Decision {
 				case CompletionIncomplete:
 					result.Incomplete = true
@@ -527,6 +590,11 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 					result.Messages = copyMessages(messages)
 					return result, nil
 				case CompletionUncertain:
+					// Observe the uncertainty signal and apply any escalation NOW:
+					// this branch continues the turn loop before the end-of-turn
+					// act point, so deferring would skip escalation entirely.
+					posture.observeUncertain()
+					applyPosture()
 					switch evaluation.Action {
 					case completionActionContinue:
 						options.Trace.Counter(trace.CounterCompletionNudges, 1)
@@ -538,13 +606,15 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 						options.Trace.Counter(trace.CounterAcceptanceChecks, 1)
 						messages = append(messages, zeroruntime.Message{
 							Role:    zeroruntime.MessageRoleUser,
-							Content: acceptanceVerificationNudge(),
+							Content: acceptanceVerificationNudge(completionContext.Objective),
 						})
 					}
 					continue
 				case CompletionComplete:
 					// Local evidence is sufficient; proceed to final diagnostics.
 				}
+			} else {
+				task.observe(taskStateEvent{kind: taskStateEventCompletion, completion: completionEvaluation{Decision: CompletionComplete}})
 			}
 			// Finalization diagnostics gate: edits from this run may still have
 			// checks in flight — the per-turn drain waits only briefly and defers
@@ -568,6 +638,11 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 		// executing so the empty-turn counter resets and plan-tracking signals
 		// stay current.
 		guards.observeTurn(collected)
+		for _, call := range collected.ToolCalls {
+			if call.Name == planToolName {
+				task.observe(taskStateEvent{kind: taskStateEventPlan, arguments: call.Arguments})
+			}
+		}
 
 		failureHint := ""
 		// turnRequestedModel records the FIRST mid-run escalation target requested
@@ -619,6 +694,7 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 			}
 			options.Trace.Counter(trace.CounterToolCalls, 1)
 			recordOutputBudgetTrace(options.Trace, toolResult)
+			task.observe(taskStateEvent{kind: taskStateEventToolResult, toolResult: toolResult})
 			if options.OnToolResult != nil {
 				options.OnToolResult(toolResult)
 			}
@@ -665,6 +741,7 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 			// would misdirect the model toward JSON shape or blocked behavior.
 			retriableFailure := isRetriableToolError(toolResult)
 			outcome := guards.observeToolResult(call.Name, retriableFailure, toolResult.Output)
+			posture.observeToolOutcome(outcome, toolResult)
 			if outcome.Stop {
 				// The assistant message advertised EVERY collected tool call, but
 				// the guard halts mid-turn so the calls after this one never run.
@@ -697,7 +774,10 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 		// the assistant's tool_results stay contiguous (a user message between
 		// tool_results breaks strict provider replay). nil SelfCorrect is a no-op.
 		if options.SelfCorrect != nil && len(changedFilesThisBatch) > 0 {
-			if feedback, _ := options.SelfCorrect.AfterEdit(ctx, dedupeStrings(changedFilesThisBatch)); feedback != "" {
+			feedback, selfCorrectOutcome := options.SelfCorrect.AfterEdit(ctx, dedupeStrings(changedFilesThisBatch))
+			posture.observeSelfCorrect(selfCorrectOutcome)
+			task.observe(taskStateEvent{kind: taskStateEventVerification, verification: selfCorrectOutcome})
+			if feedback != "" {
 				messages = append(messages, zeroruntime.Message{
 					Role:    zeroruntime.MessageRoleUser,
 					Content: feedback,
@@ -710,25 +790,53 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 		// of the run. At most one switch per turn (first request wins). A switcher
 		// error is NON-FATAL — record a brief note and continue on the current
 		// model. nil switcher ⇒ requests are ignored entirely (escalation off).
-		if turnRequestedModel != "" && options.ModelSwitcher != nil {
-			newProvider, switchErr := options.ModelSwitcher(ctx, turnRequestedModel)
+		if turnRequestedModel != "" && (options.ModelSessionSwitcher != nil || options.ModelSwitcher != nil) {
+			// Resolve the escalated model's session source. The target-aware
+			// ModelSessionSwitcher is preferred: its TurnSessionProvider carries an
+			// optimized session (and capabilities) across the swap. The legacy
+			// ModelSwitcher fallback returns a bare Provider, wrapped in the
+			// default no-op session — identical to pre-seam behavior.
+			var newSessions zeroruntime.TurnSessionProvider
+			var switchErr error
+			if options.ModelSessionSwitcher != nil {
+				newSessions, switchErr = options.ModelSessionSwitcher(ctx, turnRequestedModel)
+			} else {
+				var newProvider Provider
+				newProvider, switchErr = options.ModelSwitcher(ctx, turnRequestedModel)
+				if switchErr == nil && newProvider != nil {
+					newSessions = zeroruntime.NewProviderTurnSessionProvider(newProvider, zeroruntime.ProviderCapabilities{})
+				}
+			}
 			if switchErr != nil {
 				messages = append(messages, zeroruntime.Message{
 					Role:    zeroruntime.MessageRoleUser,
 					Content: escalationFailedNoticePrefix + " (" + turnRequestedModel + "): " + switchErr.Error() + ". Continuing on " + options.Model + ".",
 				})
-			} else if newProvider != nil {
-				// Reassign the local provider so the next turn's StreamCompletion and
-				// compaction use it; update options.Model so subsequent RunOptions.Model,
-				// context-window sizing, and usage attribution follow the new model.
-				provider = newProvider
-				options.Model = turnRequestedModel
-				options.Trace.Counter(trace.CounterModelSwitches, 1)
-				// KNOWN LIMITATION (deferred): the compactor's context-window budget
-				// is fixed at run start from options.ContextWindow and is NOT updated
-				// here, so a switch to a model with a different window keeps compacting
-				// against the original budget. Fixing it needs a ModelSwitcher contract
-				// change (return the new window) — out of scope for this change.
+			} else if newSessions != nil {
+				// Open the new session and close the old one, so the next turn's
+				// streams and compaction use the new model. For the default adapter
+				// the open never errors and Close is a no-op, so this stays
+				// byte-identical to a direct provider reassignment; the guarded open
+				// and prewarm matter once a session holds real state.
+				newSession, openErr := newSessions.OpenTurnSession(ctx)
+				if openErr != nil {
+					messages = append(messages, zeroruntime.Message{
+						Role:    zeroruntime.MessageRoleUser,
+						Content: escalationFailedNoticePrefix + " (" + turnRequestedModel + "): " + openErr.Error() + ". Continuing on " + options.Model + ".",
+					})
+				} else {
+					_ = session.Close()
+					session = newSession
+					_ = session.Prewarm(ctx)
+					provider = sessionProvider{session: session}
+					options.Model = turnRequestedModel
+					options.Trace.Counter(trace.CounterModelSwitches, 1)
+					// KNOWN LIMITATION (deferred): the compactor's context-window budget
+					// is fixed at run start from options.ContextWindow and is NOT updated
+					// here, so a switch to a model with a different window keeps compacting
+					// against the original budget. Fixing it needs the switcher to also
+					// report the new window — out of scope for this change.
+				}
 			}
 		}
 
@@ -762,6 +870,13 @@ func Run(ctx context.Context, prompt string, provider Provider, options Options)
 				Content: reminder,
 			})
 		}
+
+		// One-shot posture escalation: when an armed profile trigger fired this
+		// turn, restore the stricter knob values the profile displaced. Mutates
+		// only per-turn-read policy (the hoisted turn ceiling, reasoning effort,
+		// completion gate); never messages, model, or session. No-op forever
+		// after the first escalation and for every unprofiled run.
+		applyPosture()
 	}
 
 	if ctx.Err() != nil {
@@ -1283,10 +1398,25 @@ func executeToolCall(ctx context.Context, registry *tools.Registry, call ToolCal
 			result = registry.RebudgetAfterHook(call.Name, args, result)
 		}
 	}
+	// Stamp the sandbox risk classification for this EXECUTED call so run-policy
+	// observers can see the risk of an allowed mutation. Prefer the preflight
+	// decision's classification when the sandbox evaluated the call; otherwise
+	// run the same pure classifier the permission path uses. Denied/canceled
+	// results returned earlier keep the zero value.
+	executedRisk := sandbox.Risk{}
+	if preflightDecision != nil {
+		executedRisk = preflightDecision.Risk
+	} else if toolFound {
+		// Unknown-tool results fall through here with a nil tool; they carry a
+		// denial and keep the zero risk value.
+		executedRisk = sandbox.Classify(sandboxRequest(call.Name, tool, args, permissionGranted, permissionMode, options))
+	}
+
 	// Secret scrubbing happens at the registry boundary (the single point both
 	// the agent loop and the MCP server pass through), so result.Output is
 	// already redacted here and result.Redacted reflects whether it changed.
 	return ToolResult{
+		Risk:         executedRisk,
 		ToolCallID:   call.ID,
 		Name:         call.Name,
 		Status:       result.Status,

internal/agent/loop_test.go

@@ -8,6 +8,7 @@ import (
 	"os"
 	"os/exec"
 	"path/filepath"
+	"reflect"
 	"runtime"
 	"strings"
 	"testing"
@@ -1258,6 +1259,46 @@ func TestRunExecutesToolCallThroughRegistry(t *testing.T) {
 	}
 }
 
+func TestRunPreservesRequestPrefixAcrossTurns(t *testing.T) {
+	root := t.TempDir()
+	writeAgentTestFile(t, filepath.Join(root, "notes.txt"), "alpha\n")
+	registry := tools.NewRegistry()
+	registry.Register(tools.NewReadFileTool(root))
+	provider := &mockProvider{turns: [][]zeroruntime.StreamEvent{
+		{
+			{Type: zeroruntime.StreamEventToolCallStart, ToolCallID: "call-1", ToolName: "read_file"},
+			{Type: zeroruntime.StreamEventToolCallDelta, ToolCallID: "call-1", ArgumentsFragment: `{"path":"notes.txt"}`},
+			{Type: zeroruntime.StreamEventToolCallEnd, ToolCallID: "call-1"},
+			{Type: zeroruntime.StreamEventDone},
+		},
+		{
+			{Type: zeroruntime.StreamEventText, Content: "done"},
+			{Type: zeroruntime.StreamEventDone},
+		},
+	}}
+
+	if _, err := Run(context.Background(), "read notes", provider, Options{
+		Cwd:       root,
+		Registry:  registry,
+		SessionID: "session-stable-prefix",
+	}); err != nil {
+		t.Fatal(err)
+	}
+	if len(provider.requests) != 2 {
+		t.Fatalf("provider requests = %d, want 2", len(provider.requests))
+	}
+	first, second := provider.requests[0], provider.requests[1]
+	if len(second.Messages) < len(first.Messages) || !reflect.DeepEqual(second.Messages[:len(first.Messages)], first.Messages) {
+		t.Fatalf("first request messages must be an exact prefix of the second:\nfirst=%#v\nsecond=%#v", first.Messages, second.Messages)
+	}
+	if !reflect.DeepEqual(first.Tools, second.Tools) {
+		t.Fatalf("tool definitions drifted between turns:\nfirst=%#v\nsecond=%#v", first.Tools, second.Tools)
+	}
+	if first.PromptCacheKey != "session-stable-prefix" || second.PromptCacheKey != first.PromptCacheKey {
+		t.Fatalf("prompt cache key must remain stable: first=%q second=%q", first.PromptCacheKey, second.PromptCacheKey)
+	}
+}
+
 func TestRunSanitizesMalformedToolCallArgumentsBeforeRetry(t *testing.T) {
 	root := t.TempDir()
 	registry := tools.NewRegistry()

internal/agent/profile_controller.go

@@ -0,0 +1,118 @@
+package agent
+
+import (
+	"github.com/Gitlawb/zero/internal/sandbox"
+)
+
+// profileController observes per-turn run signals and decides at most one
+// posture escalation per run for the armed execution profile. A nil policy (or
+// a policy without Escalate) makes every method a no-op, so runs without a
+// profile are byte-identical to before — the same opt-in convention as Trace
+// and SelfCorrect.
+//
+// The controller only ever CHANGES loop policy knobs (turn ceiling, reasoning
+// effort, completion gate) through the values the loop applies at its act
+// point; it never injects messages, never touches the model or session, and
+// never interacts with the model-initiated escalate_model path.
+type profileController struct {
+	policy *PostureEscalation
+
+	escalated     bool
+	uncertainSeen int
+
+	failureTripped bool
+	riskTripped    bool
+	scTripped      bool
+	uncertainTrip  bool
+}
+
+// newProfileController is nil-safe: a nil ProfilePolicy (the default for every
+// existing caller) or a nil Escalate yields a controller that observes and
+// decides nothing.
+func newProfileController(policy *ProfilePolicy) *profileController {
+	if policy == nil {
+		return &profileController{}
+	}
+	return &profileController{policy: policy.Escalate}
+}
+
+// observeToolOutcome watches two signals from an executed tool call: the
+// repeated-failure guard's streak (reusing the guard's own counting, not a
+// second failure model) and the sandbox risk classification of an executed
+// result.
+func (c *profileController) observeToolOutcome(outcome toolFailureOutcome, result ToolResult) {
+	if c.policy == nil || c.escalated {
+		return
+	}
+	if c.policy.OnToolFailureStreak > 0 && outcome.Count >= c.policy.OnToolFailureStreak {
+		c.failureTripped = true
+	}
+	if threshold := riskRank(c.policy.OnRiskyMutation); threshold > 0 &&
+		result.DenialReason == DenialNone &&
+		riskRank(result.Risk.Level) >= threshold {
+		// The call EXECUTED (was not denied): partial failures count too, since
+		// the mutation ran. An unrecognized threshold ranks 0 and disables the
+		// signal instead of matching everything.
+		c.riskTripped = true
+	}
+}
+
+// observeUncertain counts uncertain completion evaluations (continue nudges and
+// semantic checks). Headless-only by construction: the completion gate never
+// runs interactively.
+func (c *profileController) observeUncertain() {
+	if c.policy == nil || c.escalated || c.policy.OnCompletionUncertain <= 0 {
+		return
+	}
+	c.uncertainSeen++
+	if c.uncertainSeen >= c.policy.OnCompletionUncertain {
+		c.uncertainTrip = true
+	}
+}
+
+// observeSelfCorrect watches the post-edit verification outcome. Any failing
+// outcome (correcting, reported, aborted) is a signal; passed and disabled are
+// not.
+func (c *profileController) observeSelfCorrect(outcome Outcome) {
+	if c.policy == nil || c.escalated || !c.policy.OnSelfCorrectFailure {
+		return
+	}
+	switch outcome {
+	case OutcomeCorrecting, OutcomeReported, OutcomeAborted:
+		c.scTripped = true
+	}
+}
+
+// maybeEscalate reports the escalation target exactly once, the first time any
+// armed trigger has fired. Subsequent calls return false for the rest of the
+// run: escalation is one-shot and never de-escalates, so no cooldown state is
+// needed.
+func (c *profileController) maybeEscalate() (PostureEscalation, bool) {
+	if c.policy == nil || c.escalated {
+		return PostureEscalation{}, false
+	}
+	if !(c.failureTripped || c.riskTripped || c.scTripped || c.uncertainTrip) {
+		return PostureEscalation{}, false
+	}
+	c.escalated = true
+	return *c.policy, true
+}
+
+// riskRank orders sandbox risk levels for threshold comparison. Unknown levels
+// rank 0: an unrecognized RESULT level can never meet a valid threshold, and an
+// unrecognized THRESHOLD is rejected before comparison (rank 0 disables the
+// signal) so a typo in a profile can never make every result match.
+func riskRank(level sandbox.RiskLevel) int {
+	switch level {
+	case sandbox.RiskLow:
+		return 1
+	case sandbox.RiskMedium:
+		return 2
+	case sandbox.RiskHigh:
+		return 3
+	case sandbox.RiskCritical:
+		return 4
+	default:
+		return 0
+	}
+}

internal/agent/profile_controller_test.go

@@ -0,0 +1,345 @@
+package agent
+
+import (
+	"context"
+	"strings"
+	"testing"
+
+	"github.com/Gitlawb/zero/internal/sandbox"
+	"github.com/Gitlawb/zero/internal/tools"
+	"github.com/Gitlawb/zero/internal/trace"
+	"github.com/Gitlawb/zero/internal/zeroruntime"
+)
+
+func TestProfileControllerNilPolicyIsNoOp(t *testing.T) {
+	for _, controller := range []*profileController{
+		newProfileController(nil),
+		newProfileController(&ProfilePolicy{Name: "balanced"}), // Escalate nil
+	} {
+		controller.observeToolOutcome(toolFailureOutcome{Count: 99}, ToolResult{Status: tools.StatusOK, Risk: sandbox.Risk{Level: sandbox.RiskCritical}})
+		controller.observeUncertain()
+		controller.observeSelfCorrect(OutcomeAborted)
+		if _, fired := controller.maybeEscalate(); fired {
+			t.Fatal("nil policy must never escalate")
+		}
+	}
+}
+
+func TestProfileControllerEscalatesOnFailureStreak(t *testing.T) {
+	controller := newProfileController(&ProfilePolicy{Escalate: &PostureEscalation{OnToolFailureStreak: 2, MaxTurns: 80}})
+	controller.observeToolOutcome(toolFailureOutcome{Count: 1}, ToolResult{Status: tools.StatusError})
+	if _, fired := controller.maybeEscalate(); fired {
+		t.Fatal("streak below threshold must not escalate")
+	}
+	controller.observeToolOutcome(toolFailureOutcome{Count: 2}, ToolResult{Status: tools.StatusError})
+	target, fired := controller.maybeEscalate()
+	if !fired || target.MaxTurns != 80 {
+		t.Fatalf("expected escalation with the policy target, got fired=%t target=%+v", fired, target)
+	}
+}
+
+func TestProfileControllerEscalatesOnRiskyMutation(t *testing.T) {
+	controller := newProfileController(&ProfilePolicy{Escalate: &PostureEscalation{OnRiskyMutation: sandbox.RiskHigh}})
+	// Medium risk on an executed result: below threshold.
+	controller.observeToolOutcome(toolFailureOutcome{}, ToolResult{Status: tools.StatusOK, Risk: sandbox.Risk{Level: sandbox.RiskMedium}})
+	if _, fired := controller.maybeEscalate(); fired {
+		t.Fatal("medium risk must not trip a high threshold")
+	}
+	// Critical risk on a DENIED result: never counts (the mutation never ran).
+	controller.observeToolOutcome(toolFailureOutcome{}, ToolResult{Status: tools.StatusError, DenialReason: DenialSandboxBlock, Risk: sandbox.Risk{Level: sandbox.RiskCritical}})
+	if _, fired := controller.maybeEscalate(); fired {
+		t.Fatal("a denied result must not trip the risky-mutation signal")
+	}
+	// Critical risk on an executed PARTIAL FAILURE (error status, no denial):
+	// the mutation ran, so it counts.
+	controller.observeToolOutcome(toolFailureOutcome{}, ToolResult{Status: tools.StatusError, Risk: sandbox.Risk{Level: sandbox.RiskCritical}})
+	if _, fired := controller.maybeEscalate(); !fired {
+		t.Fatal("an executed partial failure at critical risk must trip a high threshold")
+	}
+}
+
+func TestProfileControllerIgnoresInvalidRiskThreshold(t *testing.T) {
+	controller := newProfileController(&ProfilePolicy{Escalate: &PostureEscalation{OnRiskyMutation: sandbox.RiskLevel("hgih")}})
+	controller.observeToolOutcome(toolFailureOutcome{}, ToolResult{Status: tools.StatusOK, Risk: sandbox.Risk{Level: sandbox.RiskCritical}})
+	if _, fired := controller.maybeEscalate(); fired {
+		t.Fatal("an unrecognized threshold must disable the signal, not match every result")
+	}
+}
+
+func TestProfileControllerEscalatesOnSelfCorrectFailure(t *testing.T) {
+	controller := newProfileController(&ProfilePolicy{Escalate: &PostureEscalation{OnSelfCorrectFailure: true}})
+	controller.observeSelfCorrect(OutcomePassed)
+	controller.observeSelfCorrect(OutcomeDisabled)
+	if _, fired := controller.maybeEscalate(); fired {
+		t.Fatal("passing/disabled outcomes must not escalate")
+	}
+	controller.observeSelfCorrect(OutcomeReported)
+	if _, fired := controller.maybeEscalate(); !fired {
+		t.Fatal("a failing verification outcome must escalate")
+	}
+}
+
+func TestProfileControllerEscalatesOnUncertainCompletion(t *testing.T) {
+	controller := newProfileController(&ProfilePolicy{Escalate: &PostureEscalation{OnCompletionUncertain: 2}})
+	controller.observeUncertain()
+	if _, fired := controller.maybeEscalate(); fired {
+		t.Fatal("first uncertain evaluation must not trip a threshold of 2")
+	}
+	controller.observeUncertain()
+	if _, fired := controller.maybeEscalate(); !fired {
+		t.Fatal("second uncertain evaluation must escalate")
+	}
+}
+
+func TestProfileControllerEscalatesAtMostOnce(t *testing.T) {
+	controller := newProfileController(&ProfilePolicy{Escalate: &PostureEscalation{OnToolFailureStreak: 1}})
+	controller.observeToolOutcome(toolFailureOutcome{Count: 1}, ToolResult{Status: tools.StatusError})
+	if _, fired := controller.maybeEscalate(); !fired {
+		t.Fatal("expected first escalation")
+	}
+	controller.observeToolOutcome(toolFailureOutcome{Count: 5}, ToolResult{Status: tools.StatusOK, Risk: sandbox.Risk{Level: sandbox.RiskCritical}})
+	controller.observeUncertain()
+	if _, fired := controller.maybeEscalate(); fired {
+		t.Fatal("escalation must be one-shot per run")
+	}
+}
+
+// TestPostureEscalationOnUncertainCompletion proves the uncertain-path act
+// point: the completion gate's continue branch escapes the end-of-turn tail,
+// so escalation must apply before the continue. A cue turn under the gate
+// counts as uncertain, escalates, and the very next request carries the target
+// effort.
+func TestPostureEscalationOnUncertainCompletion(t *testing.T) {
+	provider := &mockProvider{turns: [][]zeroruntime.StreamEvent{
+		{
+			{Type: zeroruntime.StreamEventText, Content: "Let me read the file:"},
+			{Type: zeroruntime.StreamEventDone},
+		},
+		{
+			{Type: zeroruntime.StreamEventText, Content: "Done. All set."},
+			{Type: zeroruntime.StreamEventDone},
+		},
+	}}
+
+	recorder := trace.NewRecorder("posture-session", "run-2", "fast")
+	result, err := Run(context.Background(), "go", provider, Options{
+		Registry:                tools.NewRegistry(),
+		MaxTurns:                10,
+		RequireCompletionSignal: true,
+		Trace:                   recorder,
+		Profile: &ProfilePolicy{
+			Name: "fast",
+			Escalate: &PostureEscalation{
+				ReasoningEffort:       "high",
+				OnCompletionUncertain: 1,
+			},
+		},
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if result.FinalAnswer != "Done. All set." {
+		t.Fatalf("expected the completed answer, got %q", result.FinalAnswer)
+	}
+	if len(provider.requests) != 2 {
+		t.Fatalf("expected 2 requests, got %d", len(provider.requests))
+	}
+	if provider.requests[1].ReasoningEffort != "high" {
+		t.Fatalf("expected the post-escalation request to carry the target effort, got %q", provider.requests[1].ReasoningEffort)
+	}
+	tr := recorder.Finish()
+	if got := tr.Counter(trace.CounterPostureEscalations); got != 1 {
+		t.Fatalf("posture_escalations = %d, want 1", got)
+	}
+}
+
+// A profile that filled the run's effort displaced the provider default ("");
+// RestoreDefaultEffort lets escalation restore that default, which a plain ""
+// ReasoningEffort target cannot express (it means "leave untouched").
+func TestPostureEscalationRestoresDefaultEffort(t *testing.T) {
+	provider := &mockProvider{turns: [][]zeroruntime.StreamEvent{
+		{
+			{Type: zeroruntime.StreamEventText, Content: "Let me read the file:"},
+			{Type: zeroruntime.StreamEventDone},
+		},
+		{
+			{Type: zeroruntime.StreamEventText, Content: "Done. All set."},
+			{Type: zeroruntime.StreamEventDone},
+		},
+	}}
+
+	result, err := Run(context.Background(), "go", provider, Options{
+		Registry:                tools.NewRegistry(),
+		MaxTurns:                10,
+		ReasoningEffort:         "low",
+		RequireCompletionSignal: true,
+		Profile: &ProfilePolicy{
+			Name: "fast",
+			Escalate: &PostureEscalation{
+				RestoreDefaultEffort:  true,
+				OnCompletionUncertain: 1,
+			},
+		},
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if result.FinalAnswer != "Done. All set." {
+		t.Fatalf("expected the completed answer, got %q", result.FinalAnswer)
+	}
+	if len(provider.requests) != 2 {
+		t.Fatalf("expected 2 requests, got %d", len(provider.requests))
+	}
+	if provider.requests[0].ReasoningEffort != "low" {
+		t.Fatalf("pre-escalation request effort = %q, want the profile's low", provider.requests[0].ReasoningEffort)
+	}
+	if provider.requests[1].ReasoningEffort != "" {
+		t.Fatalf("post-escalation request effort = %q, want the restored provider default (empty)", provider.requests[1].ReasoningEffort)
+	}
+}
+
+// failingProfileTool always errors with the same signature so the repeated-
+// failure guard builds a streak.
+type failingProfileTool struct{}
+
+func (failingProfileTool) Name() string        { return "flaky_probe" }
+func (failingProfileTool) Description() string { return "always fails identically (test)" }
+func (failingProfileTool) Parameters() tools.Schema {
+	return tools.Schema{Type: "object", AdditionalProperties: false}
+}
+func (failingProfileTool) Safety() tools.Safety {
+	return tools.Safety{SideEffect: tools.SideEffectRead, Permission: tools.PermissionAllow}
+}
+func (failingProfileTool) Run(context.Context, map[string]any) tools.Result {
+	return tools.Result{Status: tools.StatusError, Output: "probe exploded: same signature"}
+}
+
+// TestPostureEscalationRaisesTurnCeilingMidRun proves the end-to-end act path:
+// a Fast-style run with a 2-turn ceiling hits a failure streak, escalates, and
+// finishes on a turn the original ceiling would have cut off, with the effort
+// target applied to the post-escalation request and the counter emitted.
+func TestPostureEscalationRaisesTurnCeilingMidRun(t *testing.T) {
+	registry := tools.NewRegistry()
+	registry.Register(failingProfileTool{})
+
+	toolTurn := []zeroruntime.StreamEvent{
+		{Type: zeroruntime.StreamEventToolCallStart, ToolCallID: "c1", ToolName: "flaky_probe"},
+		{Type: zeroruntime.StreamEventToolCallEnd, ToolCallID: "c1"},
+		{Type: zeroruntime.StreamEventDone},
+	}
+	provider := &mockProvider{turns: [][]zeroruntime.StreamEvent{
+		toolTurn,
+		toolTurn,
+		{
+			{Type: zeroruntime.StreamEventText, Content: "recovered after escalation"},
+			{Type: zeroruntime.StreamEventDone},
+		},
+	}}
+
+	recorder := trace.NewRecorder("posture-session", "run-1", "fast")
+	result, err := Run(context.Background(), "go", provider, Options{
+		Registry: registry,
+		MaxTurns: 2,
+		Trace:    recorder,
+		Profile: &ProfilePolicy{
+			Name: "fast",
+			Escalate: &PostureEscalation{
+				MaxTurns:            4,
+				ReasoningEffort:     "high",
+				OnToolFailureStreak: 2,
+			},
+		},
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if result.FinalAnswer != "recovered after escalation" {
+		t.Fatalf("expected the run to continue past the original 2-turn ceiling, got %q (turns=%d)", result.FinalAnswer, result.Turns)
+	}
+	if len(provider.requests) != 3 {
+		t.Fatalf("expected 3 requests (ceiling raised from 2 to 4), got %d", len(provider.requests))
+	}
+	if provider.requests[2].ReasoningEffort != "high" {
+		t.Fatalf("expected the post-escalation request to carry the target effort, got %q", provider.requests[2].ReasoningEffort)
+	}
+	tr := recorder.Finish()
+	if got := tr.Counter(trace.CounterPostureEscalations); got != 1 {
+		t.Fatalf("posture_escalations = %d, want 1", got)
+	}
+}
+
+// TestPostureEscalationAbsentWithoutProfile pins the no-regression invariant at
+// the loop level: the identical failing script without a profile ends at the
+// original ceiling with the max-turns answer.
+func TestPostureEscalationAbsentWithoutProfile(t *testing.T) {
+	registry := tools.NewRegistry()
+	registry.Register(failingProfileTool{})
+
+	toolTurn := []zeroruntime.StreamEvent{
+		{Type: zeroruntime.StreamEventToolCallStart, ToolCallID: "c1", ToolName: "flaky_probe"},
+		{Type: zeroruntime.StreamEventToolCallEnd, ToolCallID: "c1"},
+		{Type: zeroruntime.StreamEventDone},
+	}
+	provider := &mockProvider{turns: [][]zeroruntime.StreamEvent{toolTurn, toolTurn}}
+
+	result, err := Run(context.Background(), "go", provider, Options{
+		Registry: registry,
+		MaxTurns: 2,
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(provider.requests) > 3 {
+		t.Fatalf("unprofiled run must not extend the ceiling, got %d requests", len(provider.requests))
+	}
+	if !strings.Contains(result.FinalAnswer, "maximum number of turns") {
+		t.Fatalf("unprofiled run must end with the max-turns answer, got %q", result.FinalAnswer)
+	}
+}
+
+// TestExecuteToolCallClassifiesRiskWithoutSandbox verifies the executed-risk
+// stamp on the unsandboxed path: a read-only tool classifies low.
+func TestExecuteToolCallClassifiesRiskWithoutSandbox(t *testing.T) {
+	registry := tools.NewRegistry()
+	registry.Register(failingProfileTool{})
+
+	result, abortErr := executeToolCall(
+		context.Background(),
+		registry,
+		ToolCall{ID: "c1", Name: "flaky_probe"},
+		PermissionModeAuto,
+		Options{},
+	)
+	if abortErr != nil {
+		t.Fatal(abortErr)
+	}
+	if result.Risk.Level == "" {
+		t.Fatal("executed result must carry a risk classification")
+	}
+	if riskRank(result.Risk.Level) > riskRank(sandbox.RiskMedium) {
+		t.Fatalf("read-only tool classified %q, want at most medium", result.Risk.Level)
+	}
+}
+
+// TestUnknownToolResultCarriesZeroRisk verifies a call that never executed a
+// tool (unknown name, nil tool) keeps the zero risk value instead of panicking
+// or inventing a classification.
+func TestUnknownToolResultCarriesZeroRisk(t *testing.T) {
+	registry := tools.NewRegistry()
+	result, abortErr := executeToolCall(
+		context.Background(),
+		registry,
+		ToolCall{ID: "c1", Name: "no_such_tool"},
+		PermissionModeAuto,
+		Options{},
+	)
+	if abortErr != nil {
+		t.Fatal(abortErr)
+	}
+	if result.Status == tools.StatusOK {
+		t.Fatal("expected an error result for an unknown tool")
+	}
+	if result.Risk.Level != "" {
+		t.Fatalf("not-executed result must keep the zero risk value, got %q", result.Risk.Level)
+	}
+}

internal/agent/prompt_fingerprint.go

@@ -11,12 +11,11 @@ import (
 	"github.com/Gitlawb/zero/internal/zeroruntime"
 )
 
-// promptSubstrings are the seven cacheable sub-components of the prompt a run
-// sends to a model. Each is the literal content the sub-component would emit
-// into the system prompt or tool list, hashed so a downstream observer can
-// detect drift turn-over-turn. The substrings are produced by
-// buildPromptSubstrings and consumed by ComputePrefixFingerprint.
+// promptSubstrings are the cacheable components of the prompt a run sends to a
+// model. systemPrompt is the exact joined system-message content. The remaining
+// fields retain useful diagnostic boundaries within that prompt and tool list.
 type promptSubstrings struct {
+	systemPrompt       string
 	baseInstructions   string
 	confirmationPolicy string
 	projectContext     string
@@ -25,92 +24,55 @@ type promptSubstrings struct {
 	schema             string
 }
 
-// buildPromptSubstrings assembles the seven cacheable sub-components of the
-// prompt without joining them. The system prompt sections are produced by the
-// same private builders that buildSystemPrompt uses, so the substrings are
-// byte-identical to what would have appeared in the joined prompt. The tools
-// and schema substrings are derived from the partitioned tool list (caller-
-// provided) so this helper does not need to re-run the partition.
-//
-// This is the seam ComputePrefixFingerprint reads from. It exists as a
-// separate function (rather than returning both prompt and substrings from
-// buildSystemPrompt) so existing callers of buildSystemPrompt are unaffected
-// and the substrings helper is independently testable.
+// buildPromptSubstrings builds the prompt and retains both its exact joined
+// form and narrower diagnostic components. Callers that already built the
+// prompt should use buildPromptSubstringsFromParts to avoid repeating workspace
+// reads.
 func buildPromptSubstrings(options Options, exposed []zeroruntime.ToolDefinition) promptSubstrings {
-	core := strings.TrimSpace(options.SystemPrompt)
-	if core == "" {
-		core = strings.TrimSpace(coreSystemPrompt)
-	}
-	if core == "" {
-		core = fallbackSystemPrompt
-	}
+	return buildPromptSubstringsFromParts(buildSystemPromptParts(options), exposed)
+}
 
-	// The confirmation policy is appended unconditionally in buildSystemPrompt;
-	// the substring here is the same constant.
-	policy := strings.TrimSpace(confirmationPolicy)
-
-	// projectContext is the joined output of workspaceContext, which itself
-	// walks the AGENTS.md / ZERO.md / .zero/AGENTS.md chain. Hashing the
-	// joined string captures every guideline file's contribution and the
-	// ordering between them.
-	project := workspaceContext(options.Cwd)
-
-	// skills is the joined output of skillsContext, with each skill on its own
-	// line. The substring is the literal text the model would see.
-	skills := skillsContext(options)
-
-	// tools and schema are derived from the partitioned tool list. The list
-	// passed in is already in stable order (partitionToolsCached is the
-	// canonical stable partitioner) so concatenating in slice order is
-	// sufficient — we sort names again defensively in case a future caller
-	// hands us a list from a different partitioner.
+func buildPromptSubstringsFromParts(parts systemPromptParts, exposed []zeroruntime.ToolDefinition) promptSubstrings {
 	toolsSubstr, schemaSubstr := toolSubstrings(exposed)
 
 	return promptSubstrings{
-		baseInstructions:   core,
-		confirmationPolicy: policy,
-		projectContext:     project,
-		skills:             skills,
+		systemPrompt:       parts.prompt,
+		baseInstructions:   parts.baseInstructions,
+		confirmationPolicy: parts.confirmationPolicy,
+		projectContext:     parts.projectContext,
+		skills:             parts.skills,
 		tools:              toolsSubstr,
 		schema:             schemaSubstr,
 	}
 }
 
-// toolSubstrings extracts a stable, name-sorted digest of the partitioned tool
-// list. The tools substring is "name1\nname2\n..." (sorted) so a reordering of
-// the partition produces a different hash. The schema substring is the
-// concatenation of each tool's JSON schema in name order, so a schema change
-// is detected independently of the tool ordering.
+// toolSubstrings preserves the emitted tool order. Names and descriptions are
+// fingerprinted separately from schemas so either kind of drift is identifiable.
+// Order is part of request identity: silently sorting here would hide a provider-
+// visible reordering and could report a false cache hit.
 func toolSubstrings(exposed []zeroruntime.ToolDefinition) (toolsSubstr, schemaSubstr string) {
 	if len(exposed) == 0 {
 		return "", ""
 	}
-	names := make([]string, 0, len(exposed))
-	byName := make(map[string]zeroruntime.ToolDefinition, len(exposed))
-	for _, def := range exposed {
-		names = append(names, def.Name)
-		byName[def.Name] = def
-	}
-	sort.Strings(names)
 	var toolsSB, schemaSB strings.Builder
-	for i, name := range names {
-		if i > 0 {
-			toolsSB.WriteByte('\n')
-		}
-		toolsSB.WriteString(name)
-		def := byName[name]
+	for _, def := range exposed {
+		writeLengthPrefixed(&toolsSB, def.Name)
+		writeLengthPrefixed(&toolsSB, def.Description)
 		// Parameters is rendered to a canonical JSON string per tool. The
 		// schema-render cache in internal/agent (used by partitionToolsCached)
 		// guarantees the same render is byte-identical across turns for the
 		// same tool, so this substring is a stable hash input.
-		schemaSB.WriteString(name)
-		schemaSB.WriteByte('\n')
-		schemaSB.WriteString(canonicalSchemaString(def.Parameters))
-		schemaSB.WriteByte('\n')
+		writeLengthPrefixed(&schemaSB, def.Name)
+		writeLengthPrefixed(&schemaSB, canonicalSchemaString(def.Parameters))
 	}
 	return toolsSB.String(), schemaSB.String()
 }
 
+func writeLengthPrefixed(sb *strings.Builder, value string) {
+	fmt.Fprintf(sb, "%d:", len(value))
+	sb.WriteString(value)
+}
+
 // canonicalSchemaString renders a tool's parameter schema to a stable string.
 // tools.ToolDef.Parameters is a map[string]any in the common case; Go's
 // fmt.Sprintf("%v", m) iterates a map in random order, which would produce
@@ -227,36 +189,16 @@ func writeStable(sb *strings.Builder, v any) {
 	}
 }
 
-// ComputePrefixFingerprint returns a trace.PrefixHash (a 7-field fingerprint
-// of the prompt prefix) for one turn of a run. The seven sub-hashes are
-// independent SHA-256s of the corresponding sub-component; the complete-prefix
-// hash is a SHA-256 of the canonical concatenation of the other six, so any
-// sub-component drift is observable both individually and in aggregate.
-//
-// I/O side effects: buildPromptSubstrings calls workspaceContext, which runs
-// git (gitBranchForPrompt, FindProjectGitRoot) and reads the AGENTS.md /
-// ZERO.md / .zero/AGENTS.md chain plus the repo map. The same workspace-
-// context work is performed separately by buildSystemPrompt a few lines
-// later in the request-build path, so on every traced turn the file and
-// git reads happen twice. The duplication is intentional for this PR (a
-// follow-up will pass substrings through) and is a known perf cost only on
-// the opt-in trace path.
-//
-// A stable CompletePrefixHash across turns means the four captured
-// sub-components (baseInstructions, confirmationPolicy, projectContext,
-// skills) are byte-identical. It does NOT rule out drift in the seven
-// uncaptured sections of buildSystemPrompt (modelPromptAddendum,
-// sessionRuntimeContext, approvedCommandPrefixContext, workspaceSeedContext,
-// userGuidelines, specialistDelegationContext, responseStyleContext). For
-// default Options the four captured substrings are the full prompt; for
-// non-default Options the seven uncaptured sections may contribute, and a
-// consumer correlating CompletePrefix stability with cached_input_tokens
-// must cross-check the model_switches counter (modelPromptAddendum changes
-// on a model switch) to disambiguate "no drift" from "drift in an
-// uncaptured section."
-
+// ComputePrefixFingerprint builds and fingerprints a prompt for callers that do
+// not already have one. The agent loop uses computePrefixFingerprint with the
+// prompt parts it built once at run start, ensuring tracing observes the exact
+// request content without repeating workspace I/O.
 func ComputePrefixFingerprint(options Options, exposed []zeroruntime.ToolDefinition) prefixFingerprint {
-	subs := buildPromptSubstrings(options, exposed)
+	return computePrefixFingerprint(buildPromptSubstrings(options, exposed))
+}
+
+func computePrefixFingerprint(subs promptSubstrings) prefixFingerprint {
+	systemPrompt := sha256hex(subs.systemPrompt)
 	base := sha256hex(subs.baseInstructions)
 	policy := sha256hex(subs.confirmationPolicy)
 	project := sha256hex(subs.projectContext)
@@ -264,9 +206,10 @@ func ComputePrefixFingerprint(options Options, exposed []zeroruntime.ToolDefinit
 	toolsH := sha256hex(subs.tools)
 	schema := sha256hex(subs.schema)
 	complete := sha256hex(strings.Join([]string{
-		base, policy, project, skills, toolsH, schema,
+		systemPrompt, toolsH, schema,
 	}, "|"))
 	return prefixFingerprint{
+		SystemPromptHash:       systemPrompt,
 		BaseInstructionsHash:   base,
 		ConfirmationPolicyHash: policy,
 		ProjectContextHash:     project,
@@ -282,6 +225,7 @@ func ComputePrefixFingerprint(options Options, exposed []zeroruntime.ToolDefinit
 // in loop.go) so the trace package does not need to import this one. The field
 // names match the trace.PrefixHash JSON tags 1:1.
 type prefixFingerprint struct {
+	SystemPromptHash       string
 	BaseInstructionsHash   string
 	ConfirmationPolicyHash string
 	ProjectContextHash     string

internal/agent/prompt_fingerprint_test.go

@@ -28,38 +28,28 @@ func TestComputePrefixFingerprintStableAcrossCalls(t *testing.T) {
 	}
 }
 
-// TestComputePrefixFingerprintToolsHashIsNameSetSensitive asserts the
-// name-set sensitivity of the ToolsHash. The test name used to read as
-// "reorder changes hash" but the assertion is the opposite: the hash is
-// name-set sensitive, not name-order sensitive (toolSubstrings sorts
-// names before hashing). The partitioner is expected to produce a stable
-// order, so a name-order change is not a hash signal — a name-set change
-// is. The test asserts both: permutation of the same tools produces the
-// same hash, and adding a tool produces a different hash.
-func TestComputePrefixFingerprintToolsHashIsNameSetSensitive(t *testing.T) {
+// Tool order and descriptions are provider-visible and therefore part of the
+// cacheable prefix identity.
+func TestComputePrefixFingerprintToolsHashMatchesEmittedOrderAndDescriptions(t *testing.T) {
 	opts := Options{Cwd: t.TempDir(), SystemPrompt: "core"}
 	a := []zeroruntime.ToolDefinition{
-		{Name: "read_file", Parameters: map[string]any{"schema": "schema-read"}},
-		{Name: "grep", Parameters: map[string]any{"schema": "schema-grep"}},
+		{Name: "read_file", Description: "read", Parameters: map[string]any{"schema": "schema-read"}},
+		{Name: "grep", Description: "search", Parameters: map[string]any{"schema": "schema-grep"}},
 	}
 	b := []zeroruntime.ToolDefinition{
-		{Name: "grep", Parameters: map[string]any{"schema": "schema-grep"}},
-		{Name: "read_file", Parameters: map[string]any{"schema": "schema-read"}},
+		{Name: "grep", Description: "search", Parameters: map[string]any{"schema": "schema-grep"}},
+		{Name: "read_file", Description: "read", Parameters: map[string]any{"schema": "schema-read"}},
 	}
-	// toolSubstrings sorts names internally, so the ToolsHash is identical
-	// for a permutation of the same tools. We assert that here to document
-	// the contract: name-order independence, name-set sensitivity.
 	fpA := ComputePrefixFingerprint(opts, a)
 	fpB := ComputePrefixFingerprint(opts, b)
-	if fpA.ToolsHash != fpB.ToolsHash {
-		t.Fatalf("ToolsHash must be name-set sensitive, not name-order sensitive: a=%s b=%s", fpA.ToolsHash, fpB.ToolsHash)
+	if fpA.ToolsHash == fpB.ToolsHash {
+		t.Fatalf("ToolsHash must change when emitted order changes: a=%s b=%s", fpA.ToolsHash, fpB.ToolsHash)
 	}
-	// A different tool set must produce a different ToolsHash.
-	c := append([]zeroruntime.ToolDefinition{}, a...)
-	c = append(c, zeroruntime.ToolDefinition{Name: "bash", Parameters: map[string]any{"schema": "schema-bash"}})
+	c := append([]zeroruntime.ToolDefinition(nil), a...)
+	c[0].Description = "read a file"
 	fpC := ComputePrefixFingerprint(opts, c)
 	if fpA.ToolsHash == fpC.ToolsHash {
-		t.Fatalf("ToolsHash must change when the tool set changes: a=%s c=%s", fpA.ToolsHash, fpC.ToolsHash)
+		t.Fatalf("ToolsHash must change when a description changes: a=%s c=%s", fpA.ToolsHash, fpC.ToolsHash)
 	}
 }
 
@@ -101,25 +91,35 @@ func TestComputePrefixFingerprintSystemPromptChangeChangesBaseHash(t *testing.T)
 	}
 }
 
-// TestComputePrefixFingerprintCompleteIsAggregateOfSubHashes asserts the
-// canonical-join property: CompletePrefixHash must depend on every sub-hash.
-// This is the regression catch for "did anyone reorder or drop a sub-hash
-// from the canonical join without updating CompletePrefixHash."
+// CompletePrefixHash represents the exact joined system prompt and the ordered
+// tool definitions. Narrower system-section hashes remain diagnostic only.
 func TestComputePrefixFingerprintCompleteIsAggregateOfSubHashes(t *testing.T) {
 	opts := Options{Cwd: t.TempDir(), SystemPrompt: "core"}
 	exposed := []zeroruntime.ToolDefinition{{Name: "read_file", Parameters: map[string]any{"k": "v"}}}
 	fp := ComputePrefixFingerprint(opts, exposed)
 	// Manually compute what the canonical join should be.
 	expected := sha256hex(strings.Join([]string{
-		fp.BaseInstructionsHash,
-		fp.ConfirmationPolicyHash,
-		fp.ProjectContextHash,
-		fp.SkillsHash,
+		fp.SystemPromptHash,
 		fp.ToolsHash,
 		fp.SchemaHash,
 	}, "|"))
 	if fp.CompletePrefixHash != expected {
-		t.Fatalf("CompletePrefixHash must equal sha256hex of the canonical join of the other six:\n  got:      %s\n  expected: %s", fp.CompletePrefixHash, expected)
+		t.Fatalf("CompletePrefixHash must equal sha256hex of the exact system, tools, and schema hashes:\n  got:      %s\n  expected: %s", fp.CompletePrefixHash, expected)
+	}
+}
+
+func TestComputePrefixFingerprintCoversEverySystemPromptSection(t *testing.T) {
+	base := Options{Cwd: t.TempDir(), SystemPrompt: "core", Model: "gpt-5"}
+	changed := base
+	changed.ResponseStyle = "concise"
+
+	fpBase := ComputePrefixFingerprint(base, nil)
+	fpChanged := ComputePrefixFingerprint(changed, nil)
+	if fpBase.SystemPromptHash == fpChanged.SystemPromptHash {
+		t.Fatal("SystemPromptHash must change when any rendered system section changes")
+	}
+	if fpBase.CompletePrefixHash == fpChanged.CompletePrefixHash {
+		t.Fatal("CompletePrefixHash must include the exact full system prompt")
 	}
 }
 
@@ -141,6 +141,9 @@ func TestBuildPromptSubstringsDefaultOptions(t *testing.T) {
 	if subs.baseInstructions != "test core" {
 		t.Fatalf("baseInstructions substring must equal the core system prompt: got %q want %q", subs.baseInstructions, "test core")
 	}
+	if subs.systemPrompt != buildSystemPrompt(opts) {
+		t.Fatal("systemPrompt substring must equal the exact joined system prompt")
+	}
 	if subs.confirmationPolicy == "" {
 		t.Fatalf("confirmationPolicy substring must be non-empty (the embedded policy is always present)")
 	}
@@ -151,6 +154,9 @@ func TestBuildPromptSubstringsDefaultOptions(t *testing.T) {
 	// sha256hex of the substrings, with no truncation or reformatting
 	// between the two layers.
 	fp := ComputePrefixFingerprint(opts, nil)
+	if fp.SystemPromptHash != sha256hex(subs.systemPrompt) {
+		t.Fatalf("SystemPromptHash in fingerprint must equal sha256hex of the joined prompt")
+	}
 	if fp.BaseInstructionsHash != sha256hex(subs.baseInstructions) {
 		t.Fatalf("BaseInstructionsHash in fingerprint must equal sha256hex of the substring")
 	}

internal/agent/system_prompt.go

@@ -70,6 +70,22 @@ const (
 // guidelines), and the safety confirmation policy. It is built once per run so
 // every turn shares one (cacheable) system turn.
 func buildSystemPrompt(options Options) string {
+	return buildSystemPromptParts(options).prompt
+}
+
+// systemPromptParts retains both the exact joined prompt and the diagnostic
+// sections used by prompt-prefix tracing. Building these together prevents the
+// tracing path from re-reading workspace state and guarantees that the hash is
+// computed from the same prompt bytes placed in the request.
+type systemPromptParts struct {
+	prompt             string
+	baseInstructions   string
+	confirmationPolicy string
+	projectContext     string
+	skills             string
+}
+
+func buildSystemPromptParts(options Options) systemPromptParts {
 	core := strings.TrimSpace(options.SystemPrompt)
 	if core == "" {
 		core = strings.TrimSpace(coreSystemPrompt)
@@ -97,22 +113,31 @@ func buildSystemPrompt(options Options) string {
 	if user := userGuidelines(); user != "" {
 		sections = append(sections, user)
 	}
-	if ws := workspaceContext(options.Cwd); ws != "" {
-		sections = append(sections, ws)
+	project := workspaceContext(options.Cwd)
+	if project != "" {
+		sections = append(sections, project)
 	}
 	if delegation := specialistDelegationContext(options); delegation != "" {
 		sections = append(sections, delegation)
 	}
-	if skillsBlock := skillsContext(options); skillsBlock != "" {
+	skillsBlock := skillsContext(options)
+	if skillsBlock != "" {
 		sections = append(sections, skillsBlock)
 	}
 	if style := responseStyleContext(options); style != "" {
 		sections = append(sections, style)
 	}
-	if policy := strings.TrimSpace(confirmationPolicy); policy != "" {
+	policy := strings.TrimSpace(confirmationPolicy)
+	if policy != "" {
 		sections = append(sections, policy)
 	}
-	return strings.Join(sections, "\n\n")
+	return systemPromptParts{
+		prompt:             strings.Join(sections, "\n\n"),
+		baseInstructions:   core,
+		confirmationPolicy: policy,
+		projectContext:     project,
+		skills:             skillsBlock,
+	}
 }
 
 // responseStyleContext renders the operator-selected reply style (TUI /style) as

internal/agent/system_prompt.md

@@ -100,6 +100,9 @@ work.
 - Use exec_command with `tty: true` for interactive terminal-style commands that
   need stdin beyond Ctrl-C. `/ps` and `/stop` are user-facing TUI commands; when
   you need to clean up a running foreground command yourself, use write_stdin.
+- write_stdin's session_id is only ever an id returned by a still-running
+  exec_command; never guess or probe ids. If you have no such session, start one
+  with exec_command, or use write_file/edit_file/apply_patch for file changes.
 - write_stdin with empty input polls an existing exec_command session, and
   `\u0003` interrupts it. Sending other stdin bytes may require approval because
   it can drive the running process beyond the original command. Non-tty sessions

internal/agent/task_state.go

@@ -0,0 +1,359 @@
+package agent
+
+import (
+	"encoding/json"
+	"reflect"
+	"sort"
+	"strings"
+
+	"github.com/Gitlawb/zero/internal/tools"
+	"github.com/Gitlawb/zero/internal/trace"
+	"github.com/Gitlawb/zero/internal/zeroruntime"
+)
+
+type taskStatus string
+
+const (
+	taskStatusActive     taskStatus = "active"
+	taskStatusComplete   taskStatus = "complete"
+	taskStatusIncomplete taskStatus = "incomplete"
+)
+
+type taskPlanParity string
+
+const (
+	taskPlanParityUnknown  taskPlanParity = "unknown"
+	taskPlanParityMatch    taskPlanParity = "match"
+	taskPlanParityMismatch taskPlanParity = "mismatch"
+)
+
+type taskPlanItem struct {
+	Content string `json:"content"`
+	Status  string `json:"status"`
+	Notes   string `json:"notes,omitempty"`
+}
+
+type taskPlanState struct {
+	Items      []taskPlanItem `json:"items,omitempty"`
+	Pending    int            `json:"pending"`
+	InProgress int            `json:"in_progress"`
+	Completed  int            `json:"completed"`
+	Failed     int            `json:"failed"`
+}
+
+type taskToolState struct {
+	Succeeded int `json:"succeeded"`
+	Failed    int `json:"failed"`
+}
+
+type taskVerificationState struct {
+	Passed      int     `json:"passed"`
+	Failed      int     `json:"failed"`
+	LastOutcome Outcome `json:"last_outcome,omitempty"`
+}
+
+type taskStateSnapshot struct {
+	Revision           int                   `json:"revision"`
+	Objective          string                `json:"objective"`
+	Status             taskStatus            `json:"status"`
+	Plan               taskPlanState         `json:"plan"`
+	Tools              taskToolState         `json:"tools"`
+	Verification       taskVerificationState `json:"verification"`
+	ChangedFiles       []string              `json:"changed_files,omitempty"`
+	CompletionDecision CompletionDecision    `json:"completion_decision,omitempty"`
+	CompletionReason   string                `json:"completion_reason,omitempty"`
+	PlanParity         taskPlanParity        `json:"plan_parity"`
+}
+
+type taskStateEventKind int
+
+const (
+	taskStateEventPlan taskStateEventKind = iota + 1
+	taskStateEventToolResult
+	taskStateEventVerification
+	taskStateEventCompletion
+)
+
+type taskStateEvent struct {
+	kind         taskStateEventKind
+	arguments    string
+	toolResult   ToolResult
+	verification Outcome
+	completion   completionEvaluation
+}
+
+// taskState is a deterministic projection of facts the loop already observes.
+// It does not initiate work or replace the transcript; consumers must check
+// transcript parity before using its compact representation.
+type taskState struct {
+	snapshotValue taskStateSnapshot
+	changedFiles  map[string]struct{}
+	recorder      *trace.Recorder
+	planObserved  bool
+}
+
+func newTaskState(objective string, recorder *trace.Recorder) *taskState {
+	state := &taskState{
+		snapshotValue: taskStateSnapshot{
+			Objective:  strings.TrimSpace(objective),
+			Status:     taskStatusActive,
+			PlanParity: taskPlanParityUnknown,
+		},
+		changedFiles: map[string]struct{}{},
+	}
+	state.recorder = recorder
+	return state
+}
+
+func (state *taskState) observe(event taskStateEvent) bool {
+	if state == nil {
+		return false
+	}
+	changed := false
+	switch event.kind {
+	case taskStateEventPlan:
+		plan, ok := parseTaskPlan(event.arguments)
+		if !ok {
+			return false
+		}
+		state.snapshotValue.Plan = summarizeTaskPlan(plan)
+		state.planObserved = true
+		state.markActive()
+		changed = true
+	case taskStateEventToolResult:
+		state.markActive()
+		if event.toolResult.Status == tools.StatusOK {
+			state.snapshotValue.Tools.Succeeded++
+		} else {
+			state.snapshotValue.Tools.Failed++
+		}
+		for _, path := range event.toolResult.ChangedFiles {
+			path = strings.TrimSpace(path)
+			if path != "" {
+				state.changedFiles[path] = struct{}{}
+			}
+		}
+		state.snapshotValue.ChangedFiles = sortedKeys(state.changedFiles)
+		changed = true
+	case taskStateEventVerification:
+		if event.verification == OutcomeDisabled {
+			return false
+		}
+		state.markActive()
+		state.snapshotValue.Verification.LastOutcome = event.verification
+		switch event.verification {
+		case OutcomePassed:
+			state.snapshotValue.Verification.Passed++
+		case OutcomeCorrecting, OutcomeReported, OutcomeAborted:
+			state.snapshotValue.Verification.Failed++
+		}
+		changed = true
+	case taskStateEventCompletion:
+		state.snapshotValue.CompletionDecision = event.completion.Decision
+		state.snapshotValue.CompletionReason = event.completion.Reason
+		switch event.completion.Decision {
+		case CompletionComplete:
+			state.snapshotValue.Status = taskStatusComplete
+		case CompletionIncomplete:
+			state.snapshotValue.Status = taskStatusIncomplete
+		default:
+			state.snapshotValue.Status = taskStatusActive
+		}
+		changed = true
+	}
+	if changed {
+		state.snapshotValue.Revision++
+		if event.kind != taskStateEventToolResult {
+			state.emit()
+		}
+	}
+	return changed
+}
+
+func (state *taskState) markActive() {
+	state.snapshotValue.Status = taskStatusActive
+	state.snapshotValue.CompletionDecision = ""
+	state.snapshotValue.CompletionReason = ""
+}
+
+func (state *taskState) snapshot() taskStateSnapshot {
+	if state == nil {
+		return taskStateSnapshot{}
+	}
+	snapshot := state.snapshotValue
+	snapshot.Plan.Items = append([]taskPlanItem(nil), state.snapshotValue.Plan.Items...)
+	snapshot.ChangedFiles = append([]string(nil), state.snapshotValue.ChangedFiles...)
+	return snapshot
+}
+
+// observePlanParity compares only the latest plan projection with the plan tool
+// calls still present in messages. It mutates the snapshot and emits when the
+// parity value changes; objective, tool, and verification fields are not part of
+// this comparison.
+func (state *taskState) observePlanParity(messages []zeroruntime.Message) taskPlanParity {
+	if state == nil {
+		return taskPlanParityUnknown
+	}
+	transcriptPlan, found, valid := latestTaskPlan(messages)
+	tracked := state.snapshotValue.Plan.Items
+	parity := taskPlanParityMatch
+	if !valid || found != state.planObserved || (found && !reflect.DeepEqual(transcriptPlan, tracked)) {
+		parity = taskPlanParityMismatch
+	}
+	if state.snapshotValue.PlanParity != parity {
+		state.snapshotValue.PlanParity = parity
+		state.snapshotValue.Revision++
+		state.emit()
+	}
+	return parity
+}
+
+type completionContext struct {
+	Objective             string
+	PlanPending           bool
+	PlanMatchesTranscript bool
+}
+
+func (state *taskState) completionContext(messages []zeroruntime.Message, transcriptPlanPending bool) completionContext {
+	context := completionContext{PlanPending: transcriptPlanPending}
+	if state == nil {
+		return context
+	}
+	context.Objective = state.snapshotValue.Objective
+	context.PlanMatchesTranscript = state.observePlanParity(messages) == taskPlanParityMatch
+	if context.PlanMatchesTranscript {
+		context.PlanPending = state.snapshotValue.Plan.Pending+state.snapshotValue.Plan.InProgress > 0
+	}
+	return context
+}
+
+func (state *taskState) snapshotForCompaction(messages []zeroruntime.Message) *taskStateSnapshot {
+	if state == nil {
+		return nil
+	}
+	state.observePlanParity(messages)
+	snapshot := state.snapshot()
+	return &snapshot
+}
+
+func (state *taskState) emit() {
+	if state == nil || state.recorder == nil {
+		return
+	}
+	snapshot := state.snapshotValue
+	state.recorder.EmitTaskState(trace.TaskStateEvent{
+		Revision:            snapshot.Revision,
+		Status:              string(snapshot.Status),
+		PlanPending:         snapshot.Plan.Pending,
+		PlanInProgress:      snapshot.Plan.InProgress,
+		PlanCompleted:       snapshot.Plan.Completed,
+		PlanFailed:          snapshot.Plan.Failed,
+		ToolsSucceeded:      snapshot.Tools.Succeeded,
+		ToolsFailed:         snapshot.Tools.Failed,
+		VerificationPassed:  snapshot.Verification.Passed,
+		VerificationFailed:  snapshot.Verification.Failed,
+		VerificationOutcome: string(snapshot.Verification.LastOutcome),
+		ChangedFileCount:    len(snapshot.ChangedFiles),
+		CompletionDecision:  string(snapshot.CompletionDecision),
+		PlanParity:          string(snapshot.PlanParity),
+	})
+}
+
+func parseTaskPlan(arguments string) ([]taskPlanItem, bool) {
+	var object map[string]json.RawMessage
+	if json.Unmarshal([]byte(strings.TrimSpace(arguments)), &object) != nil {
+		return nil, false
+	}
+	rawPlan, ok := object["plan"]
+	if !ok || string(rawPlan) == "null" {
+		return nil, false
+	}
+	var parsed []struct {
+		ID      *string `json:"id"`
+		Content string  `json:"content"`
+		Status  string  `json:"status"`
+		Notes   string  `json:"notes"`
+	}
+	if json.Unmarshal(rawPlan, &parsed) != nil {
+		return nil, false
+	}
+	plan := make([]taskPlanItem, 0, len(parsed))
+	for _, raw := range parsed {
+		if raw.Content == "" {
+			return nil, false
+		}
+		plan = append(plan, taskPlanItem{
+			Content: raw.Content,
+			Status:  normalizeTaskPlanStatus(raw.Status),
+			Notes:   raw.Notes,
+		})
+	}
+	lastInProgress := -1
+	for index := range plan {
+		if plan[index].Status == "in_progress" {
+			if lastInProgress >= 0 {
+				plan[lastInProgress].Status = "completed"
+			}
+			lastInProgress = index
+		}
+	}
+	if len(plan) == 0 {
+		return nil, true
+	}
+	return plan, true
+}
+
+func normalizeTaskPlanStatus(status string) string {
+	switch strings.ToLower(strings.TrimSpace(status)) {
+	case "completed", "complete", "done", "finished", "resolved", "✓", "x", "[x]":
+		return "completed"
+	case "in_progress", "in-progress", "inprogress", "in progress", "active", "doing", "started", "current", "wip", "ongoing", "running":
+		return "in_progress"
+	case "failed", "fail", "error", "errored", "blocked", "cancelled", "canceled", "abandoned", "skipped":
+		return "failed"
+	default:
+		return "pending"
+	}
+}
+
+func summarizeTaskPlan(items []taskPlanItem) taskPlanState {
+	plan := taskPlanState{Items: append([]taskPlanItem(nil), items...)}
+	for _, item := range items {
+		switch item.Status {
+		case "completed":
+			plan.Completed++
+		case "in_progress":
+			plan.InProgress++
+		case "failed":
+			plan.Failed++
+		default:
+			plan.Pending++
+		}
+	}
+	return plan
+}
+
+func latestTaskPlan(messages []zeroruntime.Message) ([]taskPlanItem, bool, bool) {
+	for i := len(messages) - 1; i >= 0; i-- {
+		for j := len(messages[i].ToolCalls) - 1; j >= 0; j-- {
+			call := messages[i].ToolCalls[j]
+			if call.Name != planToolName {
+				continue
+			}
+			plan, ok := parseTaskPlan(call.Arguments)
+			return plan, true, ok
+		}
+	}
+	return nil, false, true
+}
+
+func sortedKeys(values map[string]struct{}) []string {
+	keys := make([]string, 0, len(values))
+	for value := range values {
+		keys = append(keys, value)
+	}
+	// Paths are evidence, not chronology. Stable ordering makes replay snapshots
+	// deterministic even if events originate from parallel read batches later.
+	sort.Strings(keys)
+	return keys
+}

internal/agent/task_state_test.go

@@ -0,0 +1,204 @@
+package agent
+
+import (
+	"context"
+	"reflect"
+	"testing"
+
+	"github.com/Gitlawb/zero/internal/tools"
+	"github.com/Gitlawb/zero/internal/trace"
+	"github.com/Gitlawb/zero/internal/zeroruntime"
+)
+
+func TestTaskStateReplayIsDeterministic(t *testing.T) {
+	events := []taskStateEvent{
+		{kind: taskStateEventPlan, arguments: `{"plan":[{"content":"inspect","status":"done"},{"content":"implement","status":"in progress"},{"content":"verify"}]}`},
+		{kind: taskStateEventToolResult, toolResult: ToolResult{Name: "apply_patch", Status: tools.StatusOK, ChangedFiles: []string{"a.go", "a.go", "b.go"}}},
+		{kind: taskStateEventToolResult, toolResult: ToolResult{Name: "go test", Status: tools.StatusError}},
+		{kind: taskStateEventVerification, verification: OutcomePassed},
+		{kind: taskStateEventCompletion, completion: completionEvaluation{Decision: CompletionUncertain, Reason: "pending work"}},
+	}
+
+	first := newTaskState("ship the change", nil)
+	second := newTaskState("ship the change", nil)
+	for _, event := range events {
+		first.observe(event)
+		second.observe(event)
+	}
+
+	if !reflect.DeepEqual(first.snapshot(), second.snapshot()) {
+		t.Fatalf("replaying the same events produced different snapshots:\nfirst:  %#v\nsecond: %#v", first.snapshot(), second.snapshot())
+	}
+	got := first.snapshot()
+	if got.Status != taskStatusActive || got.Plan.Pending != 1 || got.Plan.InProgress != 1 || got.Plan.Completed != 1 {
+		t.Fatalf("unexpected plan snapshot: %#v", got)
+	}
+	if got.Tools.Succeeded != 1 || got.Tools.Failed != 1 || len(got.ChangedFiles) != 2 {
+		t.Fatalf("unexpected evidence snapshot: %#v", got)
+	}
+	if got.Verification.Passed != 1 || got.Verification.Failed != 0 || got.Verification.LastOutcome != OutcomePassed {
+		t.Fatalf("unexpected verification snapshot: %#v", got.Verification)
+	}
+}
+
+func TestTaskStateCoalescesToolResultsIntoNextTraceEvent(t *testing.T) {
+	recorder := trace.NewRecorder("session", "run", "")
+	state := newTaskState("objective", recorder)
+	for range 20 {
+		state.observe(taskStateEvent{kind: taskStateEventToolResult, toolResult: ToolResult{Status: tools.StatusOK}})
+	}
+	if events := recorder.Finish().TaskStates; len(events) != 0 {
+		t.Fatalf("tool results should be coalesced, got %d trace events", len(events))
+	}
+
+	recorder = trace.NewRecorder("session", "run-2", "")
+	state = newTaskState("objective", recorder)
+	for range 20 {
+		state.observe(taskStateEvent{kind: taskStateEventToolResult, toolResult: ToolResult{Status: tools.StatusOK}})
+	}
+	state.observe(taskStateEvent{kind: taskStateEventCompletion, completion: completionEvaluation{Decision: CompletionComplete}})
+	events := recorder.Finish().TaskStates
+	if len(events) != 1 || events[0].ToolsSucceeded != 20 {
+		t.Fatalf("coalesced tool total missing from completion snapshot: %#v", events)
+	}
+}
+
+func TestRunEmitsTaskStateFromExistingLoopEvents(t *testing.T) {
+	registry := tools.NewRegistry()
+	registry.Register(tools.NewUpdatePlanTool())
+	provider := &mockProvider{turns: [][]zeroruntime.StreamEvent{
+		{
+			{Type: zeroruntime.StreamEventToolCallStart, ToolCallID: "plan-1", ToolName: planToolName},
+			{Type: zeroruntime.StreamEventToolCallDelta, ToolCallID: "plan-1", ArgumentsFragment: `{"plan":[{"content":"implement","status":"completed"}]}`},
+			{Type: zeroruntime.StreamEventToolCallEnd, ToolCallID: "plan-1"},
+			{Type: zeroruntime.StreamEventDone},
+		},
+		{
+			{Type: zeroruntime.StreamEventText, Content: "done"},
+			{Type: zeroruntime.StreamEventDone},
+		},
+	}}
+	recorder := trace.NewRecorder("session", "run", "")
+
+	result, err := Run(context.Background(), "implement the change", provider, Options{
+		Registry: registry,
+		Trace:    recorder,
+	})
+	if err != nil {
+		t.Fatalf("Run: %v", err)
+	}
+	if result.FinalAnswer != "done" {
+		t.Fatalf("final answer = %q, want done", result.FinalAnswer)
+	}
+	events := recorder.Finish().TaskStates
+	if len(events) < 3 {
+		t.Fatalf("expected plan, tool, completion, and parity snapshots, got %#v", events)
+	}
+	last := events[len(events)-1]
+	if last.Status != string(taskStatusComplete) || last.PlanCompleted != 1 || last.ToolsSucceeded != 1 || last.PlanParity != string(taskPlanParityMatch) {
+		t.Fatalf("unexpected final task snapshot: %#v", last)
+	}
+}
+
+func TestTaskStateSnapshotIsImmutable(t *testing.T) {
+	state := newTaskState("objective", nil)
+	state.observe(taskStateEvent{kind: taskStateEventPlan, arguments: `{"plan":[{"content":"one","status":"pending"}]}`})
+	state.observe(taskStateEvent{kind: taskStateEventToolResult, toolResult: ToolResult{Status: tools.StatusOK, ChangedFiles: []string{"one.go"}}})
+
+	snapshot := state.snapshot()
+	snapshot.Plan.Items[0].Content = "mutated"
+	snapshot.ChangedFiles[0] = "mutated.go"
+
+	fresh := state.snapshot()
+	if fresh.Plan.Items[0].Content != "one" || fresh.ChangedFiles[0] != "one.go" {
+		t.Fatalf("snapshot mutation leaked into task state: %#v", fresh)
+	}
+}
+
+func TestTaskStatePlanParityUsesLatestPlan(t *testing.T) {
+	state := newTaskState("objective", nil)
+	state.observe(taskStateEvent{kind: taskStateEventPlan, arguments: `{"plan":[{"content":"old","status":"completed"}]}`})
+	state.observe(taskStateEvent{kind: taskStateEventPlan, arguments: `{"plan":[{"content":"new","status":"in_progress"}]}`})
+
+	messages := []zeroruntime.Message{{Role: zeroruntime.MessageRoleAssistant, ToolCalls: []zeroruntime.ToolCall{
+		{Name: planToolName, Arguments: `{"plan":[{"content":"new","status":"in_progress"}]}`},
+	}}}
+	if parity := state.observePlanParity(messages); parity != taskPlanParityMatch {
+		t.Fatalf("parity = %q, want match", parity)
+	}
+
+	messages[0].ToolCalls[0].Arguments = `{"plan":[{"content":"different","status":"pending"}]}`
+	if parity := state.observePlanParity(messages); parity != taskPlanParityMismatch {
+		t.Fatalf("parity = %q, want mismatch", parity)
+	}
+}
+
+func TestTaskStateMatchesPlanToolNormalization(t *testing.T) {
+	state := newTaskState("objective", nil)
+	arguments := `{"plan":[{"content":"first","status":"in_progress"},{"content":"second","status":"in_progress"}]}`
+	state.observe(taskStateEvent{kind: taskStateEventPlan, arguments: arguments})
+
+	snapshot := state.snapshot()
+	if snapshot.Plan.Completed != 1 || snapshot.Plan.InProgress != 1 {
+		t.Fatalf("multiple active items were not normalized like the plan tool: %#v", snapshot.Plan)
+	}
+	messages := []zeroruntime.Message{{Role: zeroruntime.MessageRoleAssistant, ToolCalls: []zeroruntime.ToolCall{{Name: planToolName, Arguments: arguments}}}}
+	if parity := state.observePlanParity(messages); parity != taskPlanParityMatch {
+		t.Fatalf("normalized state should still match its transcript event, got %q", parity)
+	}
+
+	empty := newTaskState("objective", nil)
+	empty.observe(taskStateEvent{kind: taskStateEventPlan, arguments: `{"plan":[]}`})
+	emptyMessages := []zeroruntime.Message{{Role: zeroruntime.MessageRoleAssistant, ToolCalls: []zeroruntime.ToolCall{{Name: planToolName, Arguments: `{"plan":[]}`}}}}
+	if parity := empty.observePlanParity(emptyMessages); parity != taskPlanParityMatch {
+		t.Fatalf("explicit empty plan should match transcript, got %q", parity)
+	}
+}
+
+func TestTaskStateContextFallsBackWhenTranscriptDiffers(t *testing.T) {
+	state := newTaskState("objective", nil)
+	state.observe(taskStateEvent{kind: taskStateEventPlan, arguments: `{"plan":[{"content":"tracked","status":"completed"}]}`})
+	messages := []zeroruntime.Message{{Role: zeroruntime.MessageRoleAssistant, ToolCalls: []zeroruntime.ToolCall{
+		{Name: planToolName, Arguments: `{"plan":[{"content":"transcript","status":"pending"}]}`},
+	}}}
+
+	context := state.completionContext(messages, true)
+	if !context.PlanPending || context.PlanMatchesTranscript {
+		t.Fatalf("completion context must retain transcript truth on mismatch: %#v", context)
+	}
+	if context.Objective != "objective" {
+		t.Fatalf("objective lost from completion context: %#v", context)
+	}
+}
+
+func TestTaskStateCompactionSnapshotRetainsObjectiveOnPlanMismatch(t *testing.T) {
+	state := newTaskState("objective", nil)
+	state.observe(taskStateEvent{kind: taskStateEventPlan, arguments: `{"plan":[{"content":"verify","status":"pending"}]}`})
+	matching := []zeroruntime.Message{{Role: zeroruntime.MessageRoleAssistant, ToolCalls: []zeroruntime.ToolCall{
+		{Name: planToolName, Arguments: `{"plan":[{"content":"verify","status":"pending"}]}`},
+	}}}
+	if snapshot := state.snapshotForCompaction(matching); snapshot == nil || snapshot.Objective != "objective" {
+		t.Fatalf("matching transcript should produce compact state, got %#v", snapshot)
+	}
+
+	mismatching := append([]zeroruntime.Message(nil), matching...)
+	mismatching[0].ToolCalls = []zeroruntime.ToolCall{{Name: planToolName, Arguments: `{"plan":[{"content":"other","status":"pending"}]}`}}
+	if snapshot := state.snapshotForCompaction(mismatching); snapshot == nil || snapshot.Objective != "objective" || snapshot.PlanParity != taskPlanParityMismatch {
+		t.Fatalf("plan mismatch must retain immutable objective and mark mutable state uncorroborated, got %#v", snapshot)
+	}
+}
+
+func TestParseTaskPlanRejectsArgumentsTheToolWouldReject(t *testing.T) {
+	for _, arguments := range []string{
+		`{}`,
+		`{"plan":null}`,
+		`{"plan":[{"step":"alias is not accepted"}]}`,
+		`{"plan":[{"content":""}]}`,
+		`{"plan":[{"content":"valid"},{"content":""}]}`,
+		`{"plan":[{"id":4,"content":"valid"}]}`,
+	} {
+		if plan, ok := parseTaskPlan(arguments); ok {
+			t.Fatalf("parseTaskPlan(%s) = %#v, true; want rejected", arguments, plan)
+		}
+	}
+}

internal/agent/turn_session.go

@@ -0,0 +1,21 @@
+package agent
+
+import (
+	"context"
+
+	"github.com/Gitlawb/zero/internal/zeroruntime"
+)
+
+// sessionProvider adapts a TurnSession back to the Provider interface so the
+// loop's existing Provider-shaped call sites (streamWithReconnect, the
+// compaction summarizer, finalAnswerAfterMaxTurns) route their stream I/O
+// through the session without any signature change. For the default adapter
+// this reduces to the wrapped provider's StreamCompletion, so behavior is
+// byte-identical; an optimized session (PR8) takes effect here transparently.
+type sessionProvider struct {
+	session zeroruntime.TurnSession
+}
+
+func (s sessionProvider) StreamCompletion(ctx context.Context, request zeroruntime.CompletionRequest) (<-chan zeroruntime.StreamEvent, error) {
+	return s.session.Stream(ctx, request)
+}

internal/agent/turn_session_test.go

@@ -0,0 +1,367 @@
+package agent
+
+import (
+	"context"
+	"errors"
+	"strings"
+	"testing"
+
+	"github.com/Gitlawb/zero/internal/tools"
+	"github.com/Gitlawb/zero/internal/zeroruntime"
+)
+
+// fakeTurnSession wraps an inner provider for streaming while counting the
+// session lifecycle, so tests can observe exactly how the loop drives it.
+type fakeTurnSession struct {
+	inner    Provider
+	prewarms int
+	streams  int
+	closes   int
+	closeErr error
+}
+
+func (s *fakeTurnSession) Prewarm(context.Context) error { s.prewarms++; return nil }
+
+func (s *fakeTurnSession) Stream(ctx context.Context, request zeroruntime.CompletionRequest) (<-chan zeroruntime.StreamEvent, error) {
+	s.streams++
+	return s.inner.StreamCompletion(ctx, request)
+}
+
+func (s *fakeTurnSession) Compact(context.Context, zeroruntime.CompletionRequest) ([]zeroruntime.Message, error) {
+	return nil, zeroruntime.ErrCompactionUnsupported
+}
+
+func (s *fakeTurnSession) Close() error { s.closes++; return s.closeErr }
+
+// fakeTurnSessionProvider opens a scripted session (or fails to).
+type fakeTurnSessionProvider struct {
+	session *fakeTurnSession
+	openErr error
+	opens   int
+}
+
+func (p *fakeTurnSessionProvider) OpenTurnSession(context.Context) (zeroruntime.TurnSession, error) {
+	p.opens++
+	if p.openErr != nil {
+		return nil, p.openErr
+	}
+	return p.session, nil
+}
+
+func (p *fakeTurnSessionProvider) Capabilities() zeroruntime.ProviderCapabilities {
+	return zeroruntime.ProviderCapabilities{}
+}
+
+func singleAnswerTurns(answer string) [][]zeroruntime.StreamEvent {
+	return [][]zeroruntime.StreamEvent{{
+		{Type: zeroruntime.StreamEventText, Content: answer},
+		{Type: zeroruntime.StreamEventDone},
+	}}
+}
+
+// TestRunOpenTurnSessionFailureSurfaces verifies a failed session open is a
+// clean run-start error: no panic, no stream ever issued, nothing to close.
+func TestRunOpenTurnSessionFailureSurfaces(t *testing.T) {
+	positional := &mockProvider{turns: singleAnswerTurns("never reached")}
+	tsp := &fakeTurnSessionProvider{openErr: errors.New("handshake refused")}
+
+	_, err := Run(context.Background(), "go", positional, Options{
+		MaxTurns:            2,
+		TurnSessionProvider: tsp,
+	})
+	if err == nil {
+		t.Fatal("expected an error from a failed session open")
+	}
+	if !strings.Contains(err.Error(), "open turn session") || !strings.Contains(err.Error(), "handshake refused") {
+		t.Fatalf("expected a wrapped open-turn-session error, got %v", err)
+	}
+	if len(positional.requests) != 0 {
+		t.Fatalf("expected no provider request after a failed open, got %d", len(positional.requests))
+	}
+	if tsp.opens != 1 {
+		t.Fatalf("expected exactly one open attempt, got %d", tsp.opens)
+	}
+}
+
+// TestRunCloseFailureIsSafe verifies a Close error at teardown is swallowed:
+// the run still returns its normal result.
+func TestRunCloseFailureIsSafe(t *testing.T) {
+	inner := &mockProvider{turns: singleAnswerTurns("done")}
+	session := &fakeTurnSession{inner: inner, closeErr: errors.New("close blew up")}
+	tsp := &fakeTurnSessionProvider{session: session}
+
+	result, err := Run(context.Background(), "go", inner, Options{
+		MaxTurns:            2,
+		TurnSessionProvider: tsp,
+	})
+	if err != nil {
+		t.Fatalf("expected Close failure to be swallowed, got %v", err)
+	}
+	if result.FinalAnswer != "done" {
+		t.Fatalf("expected normal result despite Close error, got %q", result.FinalAnswer)
+	}
+	if session.closes != 1 {
+		t.Fatalf("expected exactly one Close at teardown, got %d", session.closes)
+	}
+}
+
+// TestRunStreamsThroughTurnSession verifies that when a TurnSessionProvider is
+// wired, every model request of the run flows through the session (and the
+// session is prewarmed once and closed once).
+func TestRunStreamsThroughTurnSession(t *testing.T) {
+	inner := &mockProvider{turns: singleAnswerTurns("via session")}
+	session := &fakeTurnSession{inner: inner}
+	tsp := &fakeTurnSessionProvider{session: session}
+
+	// The positional provider is a DIFFERENT mock: it must stay untouched,
+	// proving the session (not the positional provider) carried the run.
+	positional := &mockProvider{turns: singleAnswerTurns("wrong path")}
+
+	result, err := Run(context.Background(), "go", positional, Options{
+		MaxTurns:            2,
+		TurnSessionProvider: tsp,
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if result.FinalAnswer != "via session" {
+		t.Fatalf("expected the session-backed answer, got %q", result.FinalAnswer)
+	}
+	if len(positional.requests) != 0 {
+		t.Fatalf("expected the positional provider to be bypassed, got %d requests", len(positional.requests))
+	}
+	if session.streams == 0 || session.streams != len(inner.requests) {
+		t.Fatalf("expected every request to flow through the session: streams=%d inner=%d", session.streams, len(inner.requests))
+	}
+	if session.prewarms != 1 {
+		t.Fatalf("expected exactly one prewarm, got %d", session.prewarms)
+	}
+	if session.closes != 1 {
+		t.Fatalf("expected exactly one close, got %d", session.closes)
+	}
+}
+
+// TestRunDefaultTurnSessionMatchesRawProvider verifies the default (nil
+// TurnSessionProvider) path is behavior-identical to passing an explicit
+// default adapter over the same provider: same result, same request count.
+func TestRunDefaultTurnSessionMatchesRawProvider(t *testing.T) {
+	run := func(explicit bool) (Result, *mockProvider) {
+		provider := &mockProvider{turns: escalateThenAnswerTurns("identical")}
+		registry := tools.NewRegistry()
+		registry.Register(escalatingTool{target: ""})
+		options := Options{Registry: registry, MaxTurns: 4}
+		if explicit {
+			options.TurnSessionProvider = zeroruntime.NewProviderTurnSessionProvider(provider, zeroruntime.ProviderCapabilities{})
+		}
+		result, err := Run(context.Background(), "go", provider, options)
+		if err != nil {
+			t.Fatal(err)
+		}
+		return result, provider
+	}
+
+	defaultResult, defaultProvider := run(false)
+	explicitResult, explicitProvider := run(true)
+
+	if defaultResult.FinalAnswer != explicitResult.FinalAnswer {
+		t.Fatalf("final answers diverged: default=%q explicit=%q", defaultResult.FinalAnswer, explicitResult.FinalAnswer)
+	}
+	if defaultResult.Turns != explicitResult.Turns {
+		t.Fatalf("turn counts diverged: default=%d explicit=%d", defaultResult.Turns, explicitResult.Turns)
+	}
+	if len(defaultProvider.requests) != len(explicitProvider.requests) {
+		t.Fatalf("request counts diverged: default=%d explicit=%d", len(defaultProvider.requests), len(explicitProvider.requests))
+	}
+}
+
+// TestRunModelSwitchClosesSessionAndContinues verifies a mid-run model switch
+// closes the active session exactly once and streams the rest of the run on
+// the new provider.
+func TestRunModelSwitchClosesSessionAndContinues(t *testing.T) {
+	registry := tools.NewRegistry()
+	registry.Register(escalatingTool{target: "claude-opus-4.1"})
+
+	firstInner := &mockProvider{turns: escalateThenAnswerTurns("unused")}
+	firstSession := &fakeTurnSession{inner: firstInner}
+	tsp := &fakeTurnSessionProvider{session: firstSession}
+
+	secondProvider := &mockProvider{turns: singleAnswerTurns("answered after switch")}
+
+	result, err := Run(context.Background(), "go", firstInner, Options{
+		Registry:            registry,
+		Model:               "claude-sonnet-4.5",
+		MaxTurns:            4,
+		TurnSessionProvider: tsp,
+		ModelSwitcher: func(_ context.Context, _ string) (Provider, error) {
+			return secondProvider, nil
+		},
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if result.FinalAnswer != "answered after switch" {
+		t.Fatalf("expected the post-switch answer, got %q", result.FinalAnswer)
+	}
+	// The original session handled exactly the escalation turn, then was closed
+	// once by the swap (the teardown defer closes the NEW session, not this one).
+	if firstSession.streams != 1 {
+		t.Fatalf("expected the original session to stream exactly the escalation turn, got %d", firstSession.streams)
+	}
+	if firstSession.closes != 1 {
+		t.Fatalf("expected the swap to close the original session exactly once, got %d", firstSession.closes)
+	}
+	if len(secondProvider.requests) != 1 {
+		t.Fatalf("expected the post-switch turn on the new provider, got %d requests", len(secondProvider.requests))
+	}
+}
+
+// TestRunModelSessionSwitcherCarriesOptimizedSession verifies the target-aware
+// switcher: the switched-to TurnSessionProvider's DISTINCT session receives the
+// post-switch streams, is prewarmed exactly once, and is closed exactly once at
+// teardown — and the legacy ModelSwitcher is never consulted when both are set.
+func TestRunModelSessionSwitcherCarriesOptimizedSession(t *testing.T) {
+	registry := tools.NewRegistry()
+	registry.Register(escalatingTool{target: "claude-opus-4.1"})
+
+	firstInner := &mockProvider{turns: escalateThenAnswerTurns("unused")}
+	firstSession := &fakeTurnSession{inner: firstInner}
+	initial := &fakeTurnSessionProvider{session: firstSession}
+
+	switchedInner := &mockProvider{turns: singleAnswerTurns("answered on the switched session")}
+	switchedSession := &fakeTurnSession{inner: switchedInner}
+	switched := &fakeTurnSessionProvider{session: switchedSession}
+
+	var switchedTo string
+	legacyCalled := false
+	result, err := Run(context.Background(), "go", firstInner, Options{
+		Registry:            registry,
+		Model:               "claude-sonnet-4.5",
+		MaxTurns:            4,
+		TurnSessionProvider: initial,
+		ModelSessionSwitcher: func(_ context.Context, modelID string) (zeroruntime.TurnSessionProvider, error) {
+			switchedTo = modelID
+			return switched, nil
+		},
+		ModelSwitcher: func(_ context.Context, _ string) (Provider, error) {
+			legacyCalled = true
+			return nil, errors.New("legacy switcher must not be consulted")
+		},
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if result.FinalAnswer != "answered on the switched session" {
+		t.Fatalf("expected the switched session's answer, got %q", result.FinalAnswer)
+	}
+	if switchedTo != "claude-opus-4.1" {
+		t.Fatalf("expected switch to claude-opus-4.1, got %q", switchedTo)
+	}
+	if legacyCalled {
+		t.Fatal("legacy ModelSwitcher was consulted despite ModelSessionSwitcher being set")
+	}
+	if switched.opens != 1 {
+		t.Fatalf("expected exactly one open of the switched provider, got %d", switched.opens)
+	}
+	// The switched session carries the rest of the run: it streams the answer
+	// turn, is prewarmed exactly once at the swap, and is closed exactly once by
+	// the teardown defer.
+	if switchedSession.streams != 1 || len(switchedInner.requests) != 1 {
+		t.Fatalf("expected the switched session to stream the post-switch turn, streams=%d inner=%d", switchedSession.streams, len(switchedInner.requests))
+	}
+	if switchedSession.prewarms != 1 {
+		t.Fatalf("expected the switched session to be prewarmed exactly once, got %d", switchedSession.prewarms)
+	}
+	if switchedSession.closes != 1 {
+		t.Fatalf("expected the switched session to be closed exactly once at teardown, got %d", switchedSession.closes)
+	}
+	// The original session ends with the swap: one stream, one close.
+	if firstSession.streams != 1 || firstSession.closes != 1 {
+		t.Fatalf("expected the original session to end at the swap (streams=%d closes=%d)", firstSession.streams, firstSession.closes)
+	}
+}
+
+// TestRunModelSessionSwitcherOpenFailureIsNonFatal verifies the swap-time open
+// failure: the switcher returns a TurnSessionProvider whose OpenTurnSession
+// fails. The run stays non-fatal on the ORIGINAL session (closed only at
+// teardown), and the switch-failure note reaches the next request.
+func TestRunModelSessionSwitcherOpenFailureIsNonFatal(t *testing.T) {
+	registry := tools.NewRegistry()
+	registry.Register(escalatingTool{target: "claude-opus-4.1"})
+
+	inner := &mockProvider{turns: escalateThenAnswerTurns("recovered after failed open")}
+	session := &fakeTurnSession{inner: inner}
+	failingOpen := &fakeTurnSessionProvider{openErr: errors.New("switched handshake refused")}
+
+	result, err := Run(context.Background(), "go", inner, Options{
+		Registry:            registry,
+		Model:               "claude-sonnet-4.5",
+		MaxTurns:            4,
+		TurnSessionProvider: &fakeTurnSessionProvider{session: session},
+		ModelSessionSwitcher: func(_ context.Context, _ string) (zeroruntime.TurnSessionProvider, error) {
+			return failingOpen, nil
+		},
+	})
+	if err != nil {
+		t.Fatalf("expected a failed swap open to be non-fatal, got %v", err)
+	}
+	if result.FinalAnswer != "recovered after failed open" {
+		t.Fatalf("expected the run to recover on the original session, got %q", result.FinalAnswer)
+	}
+	if failingOpen.opens != 1 {
+		t.Fatalf("expected exactly one open attempt on the switched provider, got %d", failingOpen.opens)
+	}
+	// The original session serves both turns and is closed only at teardown —
+	// never by the failed swap.
+	if session.streams != 2 || session.closes != 1 {
+		t.Fatalf("expected the original session to serve both turns and close once (streams=%d closes=%d)", session.streams, session.closes)
+	}
+	var sawNote bool
+	for _, message := range inner.requests[1].Messages {
+		if message.Role == zeroruntime.MessageRoleUser && strings.Contains(strings.ToLower(message.Content), "could not switch") {
+			sawNote = true
+		}
+	}
+	if !sawNote {
+		t.Fatalf("expected a switch-failure note on the next turn, messages: %+v", inner.requests[1].Messages)
+	}
+}
+
+// TestRunModelSessionSwitcherErrorIsNonFatal verifies a ModelSessionSwitcher
+// error keeps the run on the current session with a transcript note — the same
+// non-fatal contract as ModelSwitcher.
+func TestRunModelSessionSwitcherErrorIsNonFatal(t *testing.T) {
+	registry := tools.NewRegistry()
+	registry.Register(escalatingTool{target: "claude-opus-4.1"})
+
+	inner := &mockProvider{turns: escalateThenAnswerTurns("recovered on original session")}
+	session := &fakeTurnSession{inner: inner}
+
+	result, err := Run(context.Background(), "go", inner, Options{
+		Registry:            registry,
+		Model:               "claude-sonnet-4.5",
+		MaxTurns:            4,
+		TurnSessionProvider: &fakeTurnSessionProvider{session: session},
+		ModelSessionSwitcher: func(_ context.Context, _ string) (zeroruntime.TurnSessionProvider, error) {
+			return nil, errors.New("session build blew up")
+		},
+	})
+	if err != nil {
+		t.Fatalf("expected session-switcher error to be non-fatal, got %v", err)
+	}
+	if result.FinalAnswer != "recovered on original session" {
+		t.Fatalf("expected the run to recover on the original session, got %q", result.FinalAnswer)
+	}
+	// Both turns stream through the original session; it closes once at teardown.
+	if session.streams != 2 || session.closes != 1 {
+		t.Fatalf("expected the original session to serve both turns and close once (streams=%d closes=%d)", session.streams, session.closes)
+	}
+	var sawNote bool
+	for _, request := range inner.requests[1].Messages {
+		if request.Role == zeroruntime.MessageRoleUser && strings.Contains(strings.ToLower(request.Content), "could not switch") {
+			sawNote = true
+		}
+	}
+	if !sawNote {
+		t.Fatalf("expected a non-fatal switch-failure note on the next turn, messages: %+v", inner.requests[1].Messages)
+	}
+}

internal/agent/types.go

@@ -75,6 +75,12 @@ type ToolResult struct {
 	// DenialReason categorizes why a tool call was blocked (empty when it ran).
 	// It lets a surface distinguish the cause precisely instead of parsing Output.
 	DenialReason DenialCategory
+	// Risk is the sandbox risk classification of this call, stamped for EXECUTED
+	// results so run-policy observers (the execution-profile controller) can see
+	// the risk level of an allowed mutation. It mirrors the classification the
+	// permission path already computes; denied or canceled results keep the zero
+	// value. Pure observation: nothing about permissions or sandboxing changes.
+	Risk sandbox.Risk
 	// LoadedTools carries the deferred-tool names a tool_search call asked the
 	// loop to expose next turn (lifted from Meta["load_tools"]). nil for every
 	// ordinary tool result; only tool_search populates it.
@@ -98,6 +104,54 @@ const (
 	DenialHookBlocked      DenialCategory = "hook_blocked"      // vetoed by a beforeTool hook
 )
 
+// ProfilePolicy is the loop-facing slice of a selected execution profile.
+// The surface (exec flag, TUI command) resolves a named profile into this
+// policy; the loop only ever sees the policy, never the catalog.
+type ProfilePolicy struct {
+	// Name labels trace counters and diagnostics (e.g. "fast"). The trace
+	// recorder's profile label is set by the caller at recorder construction.
+	Name string
+	// Escalate, when non-nil, arms one-shot in-run posture escalation.
+	Escalate *PostureEscalation
+}
+
+// PostureEscalation describes a one-shot escalation to stricter knob values,
+// applied mid-run when an armed trigger fires. Targets are the values the
+// selected profile DISPLACED at run start (i.e. "restore the balanced
+// posture"), so escalation can never introduce a value that was not already
+// valid for this run and model. Zero-valued targets leave that knob untouched.
+type PostureEscalation struct {
+	// MaxTurns raises the turn ceiling to this value when greater than the
+	// ceiling in effect. 0 leaves the ceiling untouched.
+	MaxTurns int
+	// ReasoningEffort replaces the run's effort when non-empty.
+	ReasoningEffort string
+	// RestoreDefaultEffort clears the run's effort override (back to the
+	// provider/model default) when true. It exists because the displaced value
+	// of a profile-filled effort is "" — which as a ReasoningEffort target
+	// means "leave untouched" — so restoring the default needs its own signal.
+	// Ignored when ReasoningEffort is non-empty.
+	RestoreDefaultEffort bool
+	// RestoreCompletionGate re-enables RequireCompletionSignal (headless
+	// completion semantics) when true.
+	RestoreCompletionGate bool
+
+	// Triggers. A zero value disables that signal entirely.
+	// OnToolFailureStreak fires when the repeated-failure guard observes a
+	// same-tool retriable-failure streak of at least this length.
+	OnToolFailureStreak int
+	// OnCompletionUncertain fires on the Nth uncertain completion evaluation
+	// (continue nudge or semantic check). Headless only: the completion gate
+	// never runs interactively.
+	OnCompletionUncertain int
+	// OnSelfCorrectFailure fires when a post-edit verification cycle reports a
+	// failing outcome (correcting, reported, or aborted).
+	OnSelfCorrectFailure bool
+	// OnRiskyMutation fires when an EXECUTED tool result carries a sandbox risk
+	// level at or above this threshold. Empty disables the signal.
+	OnRiskyMutation sandbox.RiskLevel
+}
+
 type PermissionRequest struct {
 	ToolCallID         string                     `json:"toolCallId"`
 	ToolName           string                     `json:"name"`
@@ -283,6 +337,27 @@ type Options struct {
 	// request), so every existing caller is unaffected. A returned error is
 	// non-fatal: the run continues on the current model.
 	ModelSwitcher func(ctx context.Context, modelID string) (Provider, error)
+	// TurnSessionProvider, when set, supplies the turn session the run streams
+	// through — the seam an optimized provider session (connection reuse,
+	// prewarm, native compaction) plugs into without touching the loop. nil
+	// keeps the default: the loop wraps the passed provider in a no-op session
+	// whose Stream IS provider.StreamCompletion, so behavior is byte-identical
+	// and every existing caller is unaffected.
+	TurnSessionProvider zeroruntime.TurnSessionProvider
+	// ModelSessionSwitcher, when set, is the target-aware escalation hook: the
+	// loop prefers it over ModelSwitcher, and its TurnSessionProvider keeps an
+	// optimized session (and its capabilities) across a mid-run model switch.
+	// nil falls back to ModelSwitcher, whose bare Provider is wrapped in the
+	// default no-op session — today's behavior, unchanged. Same non-fatal error
+	// contract as ModelSwitcher: a returned error records a note and the run
+	// continues on the current model and session.
+	ModelSessionSwitcher func(ctx context.Context, modelID string) (zeroruntime.TurnSessionProvider, error)
+	// Profile, when set, arms the execution-profile posture controller for this
+	// run (auto-escalation to stricter knob values on failure/uncertainty/risky
+	// mutation signals). nil — the default everywhere today — leaves the loop
+	// byte-identical: no observation, no escalation, no counters. Same opt-in
+	// convention as Trace and SelfCorrect.
+	Profile *ProfilePolicy
 	// Trace, when set, records per-turn timing for the run: the loop stamps
 	// spans (prompt build, generation, tool execution, permission wait,
 	// compaction, provider connect) and counters (model requests, tool calls,

internal/cli/app.go

@@ -1341,6 +1341,9 @@ Flags:
       --spec-reasoning-effort <effort>
                                     Override draft reasoning effort when --use-spec is set
       --max-turns <number>           Override the maximum agent loop turns
+      --exec-profile <name>          Apply an execution profile (balanced, fast, thorough): loop
+                                    posture only (turn budget, effort, self-correction, escalation);
+                                    composes with --mode (which picks the model) and explicit flags win
       --auto <low|medium|high>       Set exec autonomy; high enables unsafe tools
       --enabled-tools <tools>        Only expose these comma or space separated tools
       --disabled-tools <tools>       Hide these comma or space separated tools

internal/cli/exec.go

@@ -13,6 +13,7 @@ import (
 	"github.com/Gitlawb/zero/internal/agent"
 	"github.com/Gitlawb/zero/internal/config"
 	"github.com/Gitlawb/zero/internal/errhint"
+	"github.com/Gitlawb/zero/internal/execprofile"
 	"github.com/Gitlawb/zero/internal/imageinput"
 	"github.com/Gitlawb/zero/internal/lsp"
 	"github.com/Gitlawb/zero/internal/modelregistry"
@@ -69,7 +70,14 @@ type execOptions struct {
 	// intentionally inert: nothing consumes it. Model selection is driven by
 	// --model / --mode instead. See writeExecHelp ("Accept legacy model profile
 	// selection") and TestRunExecAcceptsLegacyModelProfileFlags.
-	modelProfile          string
+	modelProfile string
+	// execProfile selects a named execution profile (balanced, fast, thorough):
+	// a loop-posture bundle (turn budget, reasoning effort, self-correction,
+	// escalation triggers) applied AFTER --mode with the same
+	// fill-only-if-unset rule, so precedence is explicit flag > mode > profile.
+	// Distinct from the mode preset also named "fast" (which picks a model) and
+	// from the legacy inert --profile above. See internal/execprofile.
+	execProfile           string
 	reasoningEffort       string
 	useSpec               bool
 	specModel             string
@@ -167,6 +175,14 @@ func runExec(args []string, stdout io.Writer, stderr io.Writer, deps appDeps) in
 	if err := applyExecMode(&options); err != nil {
 		return writeExecFormatUsageError(stdout, stderr, options.outputFormat, err.Error())
 	}
+	// A profile tunes loop posture the way a mode tunes model selection, and
+	// runs after it so the mode's fills count as "set" and win. MaxTurns is
+	// deferred to config resolution below, where the displaced resolved budget
+	// is known and becomes the escalation restore target.
+	execProfile, execProfileFilledEffort, err := applyExecProfile(&options)
+	if err != nil {
+		return writeExecFormatUsageError(stdout, stderr, options.outputFormat, err.Error())
+	}
 
 	workspaceRoot, err := resolveWorkspaceRoot(options.cwd, deps)
 	if err != nil {
@@ -272,6 +288,8 @@ func runExec(args []string, stdout io.Writer, stderr io.Writer, deps appDeps) in
 		}
 		return writeExecProviderError(stdout, stderr, options.outputFormat, "provider_error", err.Error())
 	}
+	var displacedMaxTurns int
+	resolved.MaxTurns, displacedMaxTurns = applyProfileTurnBudget(execProfile, options.maxTurns, resolved.MaxTurns)
 	registerLocalControlTools(registry, workspaceRoot, resolved.LocalControl)
 	if err := validateExecToolFilters(options, registry); err != nil {
 		return writeExecFormatUsageError(stdout, stderr, options.outputFormat, err.Error())
@@ -412,6 +430,36 @@ func runExec(args []string, stdout io.Writer, stderr io.Writer, deps appDeps) in
 		}
 	}
 
+	// Optimized OpenAI turn sessions (ZERO_OPENAI_TURN_SESSION, default off).
+	// nil when gated off or the profile is ineligible: agent.Run then wraps the
+	// provider in its default adapter — the exact code path of today. The
+	// session switcher is installed only when the run START is optimized, so
+	// the legacy ModelSwitcher path above stays untouched otherwise.
+	turnSessions, _ := providers.OptimizedTurnSessions(resolved.Provider, provider, providers.Options{})
+	var modelSessionSwitcher func(context.Context, string) (zeroruntime.TurnSessionProvider, error)
+	if options.allowEscalation && turnSessions != nil {
+		modelSessionSwitcher = func(_ context.Context, modelID string) (zeroruntime.TurnSessionProvider, error) {
+			switchedProfile := resolved.Provider
+			switchedProfile.Model = modelID
+			switchedProvider, err := deps.newProvider(switchedProfile)
+			if err != nil {
+				return nil, err
+			}
+			if switchedProvider == nil {
+				// The loop treats a nil session source as "no swap" — mirror the
+				// legacy closure's (nil, nil) contract.
+				return nil, nil
+			}
+			currentModel = modelID
+			if optimized, ok := providers.OptimizedTurnSessions(switchedProfile, switchedProvider, providers.Options{}); ok {
+				return optimized, nil
+			}
+			// Ineligible switch target: default adapter, but with the switched
+			// model's resolved capability projection preserved.
+			return providers.DefaultTurnSessions(switchedProfile, switchedProvider, providers.Options{}), nil
+		}
+	}
+
 	runMetadata, err := resolveExecRunMetadata(resolved.Provider)
 	if err != nil {
 		return writeExecProviderError(stdout, stderr, options.outputFormat, "provider_error", err.Error())
@@ -456,6 +504,13 @@ func runExec(args []string, stdout io.Writer, stderr io.Writer, deps appDeps) in
 			reasoningEffort:    forwardEffort,
 			specPermissionMode: permissionMode,
 			notifier:           notifier,
+			// The profile displaced resolved.MaxTurns above, so the spec-draft
+			// run arms the same escalation policy as the main run (in practice
+			// only the failure-streak and risky-mutation triggers can fire
+			// here: the draft runs without the completion gate or a
+			// self-corrector).
+			profilePolicy: execProfile.Policy(displacedMaxTurns,
+				specProfileEffortFilled(execProfileFilledEffort, options.specReasoningEffort)),
 		})
 	}
 
@@ -498,7 +553,7 @@ func runExec(args []string, stdout io.Writer, stderr io.Writer, deps appDeps) in
 	var traceRecorder *trace.Recorder
 	var traceSnapshot *trace.TurnTrace
 	if tracePath != "" {
-		traceRecorder = trace.NewRecorder(preparedSession.Session.SessionID, runID, "")
+		traceRecorder = trace.NewRecorder(preparedSession.Session.SessionID, runID, execProfile.Name)
 		defer func() {
 			if err := writeTraceSnapshot(traceSnapshot, tracePath, stderr); err != nil {
 				fmt.Fprintf(stderr, "[zero] failed to write trace: %s\n", err)
@@ -572,29 +627,32 @@ func runExec(args []string, stdout io.Writer, stderr io.Writer, deps appDeps) in
 	hookDispatcher, hookSkip := newHookDispatcherWithExtra(workspaceRoot, pluginActivation.hooks, trustRoot)
 	emitTrustNotice(stderr, hookSkip, pluginActivation.trustSkip, mcpSkip)
 	result, err := agent.Run(runCtx, agentPrompt, provider, agent.Options{
-		MaxTurns:         resolved.MaxTurns,
-		ContextWindow:    resolveAgentContextWindow(runCtx, modelRegistry, resolved.Provider),
-		DeferThreshold:   effectiveDeferThreshold,
-		Specialists:      specialistRuntime.specialistInfos(),
-		Skills:           pluginActivation.skillInfos(deps.skillsDir()),
-		SessionID:        preparedSession.Session.SessionID,
-		CallingSessionID: options.callingSessionID,
-		CallingToolUseID: options.callingToolUseID,
-		Tag:              options.tag,
-		Depth:            options.depth,
-		SessionTitle:     sessionTitle,
-		ProviderName:     resolved.Provider.Name,
-		Model:            resolved.Provider.Model,
-		ModelSwitcher:    modelSwitcher,
-		ReasoningEffort:  forwardEffort,
-		Trace:            traceRecorder,
-		Cwd:              workspaceRoot,
-		Images:           images,
-		Registry:         registry,
-		PermissionMode:   permissionMode,
-		Autonomy:         options.autonomy,
-		SelfCorrect:      selfCorrector,
-		FileDiagnostics:  fileDiagnostics,
+		MaxTurns:             resolved.MaxTurns,
+		ContextWindow:        resolveAgentContextWindow(runCtx, modelRegistry, resolved.Provider),
+		DeferThreshold:       effectiveDeferThreshold,
+		Specialists:          specialistRuntime.specialistInfos(),
+		Skills:               pluginActivation.skillInfos(deps.skillsDir()),
+		SessionID:            preparedSession.Session.SessionID,
+		CallingSessionID:     options.callingSessionID,
+		CallingToolUseID:     options.callingToolUseID,
+		Tag:                  options.tag,
+		Depth:                options.depth,
+		SessionTitle:         sessionTitle,
+		ProviderName:         resolved.Provider.Name,
+		Model:                resolved.Provider.Model,
+		ModelSwitcher:        modelSwitcher,
+		TurnSessionProvider:  turnSessions,
+		ModelSessionSwitcher: modelSessionSwitcher,
+		ReasoningEffort:      forwardEffort,
+		Trace:                traceRecorder,
+		Cwd:                  workspaceRoot,
+		Images:               images,
+		Registry:             registry,
+		PermissionMode:       permissionMode,
+		Autonomy:             options.autonomy,
+		SelfCorrect:          selfCorrector,
+		FileDiagnostics:      fileDiagnostics,
+		Profile:              execProfile.Policy(displacedMaxTurns, execProfileFilledEffort),
 		// Headless exec: don't accept a no-tool-call turn as "done" while work
 		// clearly remains (pending plan items / a mid-step continuation cue) —
 		// nudge to continue, and finalize as INCOMPLETE rather than false success
@@ -881,6 +939,12 @@ func providerSensitiveEnvKeys(resolved config.ResolvedConfig) []string {
 // applyConfiguredSandboxPolicy overlays every config-sourced sandbox knob onto
 // the default policy.
 func applyConfiguredSandboxPolicy(policy sandbox.Policy, cfg config.SandboxConfig) sandbox.Policy {
+	// An explicit `"sandbox": {"enabled": false}` disables the sandbox: every
+	// request is allowed and commands run unwrapped. Only reachable from global
+	// config / CLI — project config never sets Enabled (see mergeProjectConfig).
+	if cfg.Enabled != nil && !*cfg.Enabled {
+		policy.Mode = sandbox.ModeDisabled
+	}
 	if network := strings.TrimSpace(cfg.Network); network != "" {
 		switch sandbox.NetworkMode(network) {
 		case sandbox.NetworkAllow, sandbox.NetworkDeny:
@@ -1101,6 +1165,67 @@ func applyExecMode(options *execOptions) error {
 	return nil
 }
 
+// applyExecProfile expands an --exec-profile selection onto the exec options.
+// Like applyExecMode it only fills fields the caller left unset, and it runs
+// AFTER the mode so a mode-filled field counts as set: precedence is explicit
+// flag > --mode > --exec-profile. Only the options-level knobs are applied here
+// (reasoning effort still flows through the model-supported gating downstream;
+// self-correct is presence-only so a profile can only turn it on). MaxTurns is
+// intentionally NOT applied here — the profile displaces the resolved budget
+// after config resolution, where the displaced value is known and becomes the
+// escalation restore target. An unknown profile is a usage error listing the
+// valid names. Not to be confused with the legacy inert --profile flag.
+// The second return reports whether the profile actually filled the reasoning
+// effort (false when the user set one explicitly), so the escalation policy
+// knows the displaced effort was the provider default and can restore it.
+func applyExecProfile(options *execOptions) (execprofile.Profile, bool, error) {
+	name := strings.TrimSpace(options.execProfile)
+	if name == "" {
+		return execprofile.Profile{}, false, nil
+	}
+	profile, ok := execprofile.Lookup(name)
+	if !ok {
+		return execprofile.Profile{}, false, execUsageError{fmt.Sprintf("unknown execution profile %q. Valid profiles: %s.", options.execProfile, strings.Join(execprofile.Names(), ", "))}
+	}
+	effortFilled := false
+	if options.reasoningEffort == "" && profile.ReasoningEffort != "" {
+		options.reasoningEffort = profile.ReasoningEffort
+		effortFilled = true
+	}
+	// Spec-safe projection: the spec-draft path wires no self-corrector (an
+	// explicit --self-correct --use-spec is rejected at parse time, before
+	// profiles apply), so a profile's self-correct knob is meaningless there.
+	// Project it away rather than erroring: the user asked for a thorough
+	// DRAFT, and the profile's other knobs (budget, effort) apply to it fine.
+	if profile.SelfCorrect && !options.useSpec {
+		options.selfCorrect = true
+	}
+	return profile, effortFilled, nil
+}
+
+// specProfileEffortFilled reports whether the profile's effort fill actually
+// governs the spec-draft run. An explicit --spec-reasoning-effort replaces the
+// filled effort for the draft, so the escalation's effort restore must not arm
+// there: escalation must never clear an effort the user pinned by hand.
+func specProfileEffortFilled(effortFilled bool, specReasoningEffort string) bool {
+	return effortFilled && strings.TrimSpace(specReasoningEffort) == ""
+}
+
+// applyProfileTurnBudget decides the run's turn budget once config is resolved.
+// The profile displaces the RESOLVED value, not the flag: a pinned budget
+// always wins and the profile backs off entirely, while an env/config budget
+// is the "balanced posture" a mid-run escalation can restore. pinnedMaxTurns
+// is any caller-pinned budget — an explicit --max-turns flag OR a mode
+// preset's fill (both land in options.maxTurns, and a mode outranks a profile
+// just like the flag does). displaced is 0 when nothing was displaced, so an
+// escalation leaves the ceiling untouched.
+func applyProfileTurnBudget(profile execprofile.Profile, pinnedMaxTurns int, resolvedMaxTurns int) (effective int, displaced int) {
+	if profile.MaxTurns > 0 && pinnedMaxTurns == 0 {
+		return profile.MaxTurns, resolvedMaxTurns
+	}
+	return resolvedMaxTurns, 0
+}
+
 // resolveSelectedModel routes a user-supplied --model value through the model
 // registry so that fuzzy aliases (e.g. "sonnet 4.5") resolve to canonical ids
 // and deprecated models auto-redirect to their fallback. It returns the model id

internal/cli/exec_parse.go

@@ -147,6 +147,19 @@ func parseExecArgs(args []string) (execOptions, bool, error) {
 			index = next
 		case strings.HasPrefix(arg, "--profile="):
 			options.modelProfile = strings.TrimSpace(strings.TrimPrefix(arg, "--profile="))
+		case arg == "--exec-profile":
+			value, next, err := nextFlagValue(args, index, arg)
+			if err != nil {
+				return options, false, err
+			}
+			options.execProfile = strings.TrimSpace(value)
+			index = next
+		case strings.HasPrefix(arg, "--exec-profile="):
+			value, err := requiredInlineFlagValue(arg, "--exec-profile")
+			if err != nil {
+				return options, false, err
+			}
+			options.execProfile = value
 		case arg == "-r" || arg == "--reasoning-effort":
 			value, next, err := nextFlagValue(args, index, arg)
 			if err != nil {

internal/cli/exec_profile_test.go

@@ -0,0 +1,224 @@
+package cli
+
+import (
+	"os"
+	"path/filepath"
+	"reflect"
+	"strings"
+	"testing"
+
+	"github.com/Gitlawb/zero/internal/execprofile"
+)
+
+func TestApplyExecProfileFillsOnlyUnset(t *testing.T) {
+	// Unset effort: the profile fills it and reports the fill, so the
+	// escalation policy knows the displaced effort was the provider default.
+	options := execOptions{execProfile: "fast"}
+	profile, effortFilled, err := applyExecProfile(&options)
+	if err != nil {
+		t.Fatalf("applyExecProfile: %v", err)
+	}
+	if profile.Name != "fast" {
+		t.Fatalf("profile.Name = %q, want fast", profile.Name)
+	}
+	if options.reasoningEffort != "low" {
+		t.Fatalf("reasoningEffort = %q, want the profile's low", options.reasoningEffort)
+	}
+	if !effortFilled {
+		t.Fatal("the profile filled the effort, so effortFilled must report it")
+	}
+
+	// Set effort (explicit flag or a mode's fill): the profile backs off.
+	options = execOptions{execProfile: "fast", reasoningEffort: "high"}
+	_, effortFilled, err = applyExecProfile(&options)
+	if err != nil {
+		t.Fatalf("applyExecProfile: %v", err)
+	}
+	if options.reasoningEffort != "high" {
+		t.Fatalf("reasoningEffort = %q, an explicit value must win over the profile", options.reasoningEffort)
+	}
+	if effortFilled {
+		t.Fatal("the profile backed off, so effortFilled must be false")
+	}
+}
+
+func TestApplyExecProfileThoroughArmsSelfCorrect(t *testing.T) {
+	options := execOptions{execProfile: "thorough"}
+	if _, _, err := applyExecProfile(&options); err != nil {
+		t.Fatalf("applyExecProfile: %v", err)
+	}
+	if !options.selfCorrect {
+		t.Fatal("thorough must arm the post-edit self-correct loop")
+	}
+	if options.reasoningEffort != "high" {
+		t.Fatalf("reasoningEffort = %q, want high", options.reasoningEffort)
+	}
+}
+
+// The spec-draft path wires no self-corrector (explicit --self-correct
+// --use-spec is rejected at parse time, before profiles apply), so a profile's
+// self-correct knob is projected away under --use-spec instead of silently
+// arming a knob the draft runner ignores. The other knobs still apply.
+func TestApplyExecProfileSpecProjectsAwaySelfCorrect(t *testing.T) {
+	options := execOptions{execProfile: "thorough", useSpec: true}
+	if _, _, err := applyExecProfile(&options); err != nil {
+		t.Fatalf("applyExecProfile: %v", err)
+	}
+	if options.selfCorrect {
+		t.Fatal("thorough under --use-spec must not arm self-correct (the draft runner has none)")
+	}
+	if options.reasoningEffort != "high" {
+		t.Fatalf("reasoningEffort = %q, the profile's other knobs must still apply", options.reasoningEffort)
+	}
+}
+
+func TestApplyExecProfileUnknownIsUsageError(t *testing.T) {
+	options := execOptions{execProfile: "turbo"}
+	_, _, err := applyExecProfile(&options)
+	if err == nil {
+		t.Fatal("expected a usage error for an unknown profile")
+	}
+	if _, ok := err.(execUsageError); !ok {
+		t.Fatalf("expected execUsageError, got %T: %v", err, err)
+	}
+	for _, name := range execprofile.Names() {
+		if !strings.Contains(err.Error(), name) {
+			t.Fatalf("usage error must list %q, got %q", name, err.Error())
+		}
+	}
+}
+
+// Precedence: explicit flag > --mode > --exec-profile. The deep mode fills
+// effort=high and max-turns=160; the fast profile must not override either,
+// because a mode-filled field counts as set by the time the profile runs.
+func TestExecProfilePrecedenceFlagOverModeOverProfile(t *testing.T) {
+	options := execOptions{mode: "deep", execProfile: "fast"}
+	if err := applyExecMode(&options); err != nil {
+		t.Fatalf("applyExecMode: %v", err)
+	}
+	profile, _, err := applyExecProfile(&options)
+	if err != nil {
+		t.Fatalf("applyExecProfile: %v", err)
+	}
+	if options.reasoningEffort != "high" {
+		t.Fatalf("reasoningEffort = %q, the mode's high must win over the profile's low", options.reasoningEffort)
+	}
+	// The mode filled options.maxTurns, so the turn budget must not be displaced.
+	effective, displaced := applyProfileTurnBudget(profile, options.maxTurns, options.maxTurns)
+	if effective != 160 || displaced != 0 {
+		t.Fatalf("turn budget = (%d, displaced %d), the mode's 160 must win with nothing displaced", effective, displaced)
+	}
+}
+
+func TestApplyProfileTurnBudget(t *testing.T) {
+	fast, _ := execprofile.Lookup("fast")
+	balanced, _ := execprofile.Lookup("balanced")
+
+	// No explicit flag: the profile displaces the resolved budget.
+	if effective, displaced := applyProfileTurnBudget(fast, 0, 80); effective != 30 || displaced != 80 {
+		t.Fatalf("fast over resolved 80 = (%d, %d), want (30, 80)", effective, displaced)
+	}
+	// Explicit --max-turns pins the budget; the profile backs off entirely.
+	if effective, displaced := applyProfileTurnBudget(fast, 50, 50); effective != 50 || displaced != 0 {
+		t.Fatalf("fast with explicit 50 = (%d, %d), want (50, 0)", effective, displaced)
+	}
+	// Balanced never displaces anything.
+	if effective, displaced := applyProfileTurnBudget(balanced, 0, 80); effective != 80 || displaced != 0 {
+		t.Fatalf("balanced over resolved 80 = (%d, %d), want (80, 0)", effective, displaced)
+	}
+}
+
+// The no-regression invariant at the options level: selecting balanced leaves
+// the options byte-identical to not selecting a profile at all, and produces
+// no loop policy.
+func TestExecProfileBalancedLeavesOptionsUntouched(t *testing.T) {
+	options := execOptions{execProfile: "balanced"}
+	reference := options
+	profile, _, err := applyExecProfile(&options)
+	if err != nil {
+		t.Fatalf("applyExecProfile: %v", err)
+	}
+	if !reflect.DeepEqual(options, reference) {
+		t.Fatalf("balanced changed the options: %+v vs %+v", options, reference)
+	}
+	if policy := profile.Policy(80, false); policy != nil {
+		t.Fatalf("balanced must produce a nil policy, got %+v", policy)
+	}
+}
+
+// --exec-profile= with an inline empty value must error like its two-token
+// form and the sibling --mode=, not silently run the default posture (the
+// classic shell hazard: an unset variable expanding into the inline form).
+func TestParseExecProfileInlineEmptyIsError(t *testing.T) {
+	_, _, err := parseExecArgs([]string{"--exec-profile=", "hello"})
+	if err == nil || !strings.Contains(err.Error(), "--exec-profile requires a value") {
+		t.Fatalf("expected a required-value error, got %v", err)
+	}
+}
+
+// Full echo-provider run: the selected profile must be stamped into the
+// per-turn trace so benchmark attribution can group runs by posture.
+func TestRunExecTraceRecordsSelectedProfile(t *testing.T) {
+	tracePath := filepath.Join(t.TempDir(), "trace.ndjson")
+	exitCode, _, stderr := runExecWithEcho(t, []string{
+		"exec", "--exec-profile", "fast", "--trace", tracePath, "hello",
+	})
+	if exitCode != exitSuccess {
+		t.Fatalf("expected exit %d, got %d: %s", exitSuccess, exitCode, stderr)
+	}
+	raw, err := os.ReadFile(tracePath)
+	if err != nil {
+		t.Fatalf("read trace: %v", err)
+	}
+	if !strings.Contains(string(raw), `"profile":"fast"`) {
+		t.Fatalf("trace must record the selected profile, got %s", raw)
+	}
+}
+
+// An explicit --spec-reasoning-effort governs the spec draft's effort, so the
+// escalation effort-restore must not arm there even when the profile filled
+// the (unused) main effort.
+func TestSpecProfileEffortFilled(t *testing.T) {
+	if !specProfileEffortFilled(true, "") {
+		t.Fatal("no explicit spec effort: the profile's fill governs, restore must arm")
+	}
+	if specProfileEffortFilled(true, "high") {
+		t.Fatal("an explicit spec effort must disarm the effort restore")
+	}
+	if specProfileEffortFilled(false, "") {
+		t.Fatal("nothing filled, nothing to restore")
+	}
+}
+
+// The trace label deliberately records the SELECTED profile, balanced
+// included, so captures are self-describing. Run behavior stays identical
+// (balanced fills nothing and arms no policy); the label is the one intended
+// difference in the opt-in trace artifact.
+func TestRunExecTraceRecordsBalancedSelection(t *testing.T) {
+	tracePath := filepath.Join(t.TempDir(), "trace.ndjson")
+	exitCode, _, stderr := runExecWithEcho(t, []string{
+		"exec", "--exec-profile", "balanced", "--trace", tracePath, "hello",
+	})
+	if exitCode != exitSuccess {
+		t.Fatalf("expected exit %d, got %d: %s", exitSuccess, exitCode, stderr)
+	}
+	raw, err := os.ReadFile(tracePath)
+	if err != nil {
+		t.Fatalf("read trace: %v", err)
+	}
+	if !strings.Contains(string(raw), `"profile":"balanced"`) {
+		t.Fatalf("trace must record the balanced selection, got %s", raw)
+	}
+}
+
+func TestRunExecUnknownExecProfileIsUsageError(t *testing.T) {
+	exitCode, _, stderr := runExecWithEcho(t, []string{
+		"exec", "--exec-profile", "turbo", "hello",
+	})
+	if exitCode != exitUsage {
+		t.Fatalf("expected usage exit %d, got %d", exitUsage, exitCode)
+	}
+	if !strings.Contains(stderr, "unknown execution profile") {
+		t.Fatalf("expected an unknown-profile usage error, got %q", stderr)
+	}
+}

internal/cli/exec_sessions.go

@@ -72,8 +72,11 @@ func preflightExecSession(options execOptions) error {
 		if session == nil {
 			return execUsageError{"Zero session not found: " + options.resume}
 		}
+		if !sessions.IsResumableKind(session.SessionKind) {
+			return execUsageError{"Zero session is not resumable: " + options.resume}
+		}
 	case options.resumeLatest:
-		latest, err := store.Latest()
+		latest, err := store.LatestResumable()
 		if err != nil {
 			return err
 		}

internal/cli/exec_sessions_test.go

@@ -0,0 +1,26 @@
+package cli
+
+import (
+	"strings"
+	"testing"
+
+	"github.com/Gitlawb/zero/internal/sessions"
+)
+
+func TestPreflightRejectsSideSessions(t *testing.T) {
+	t.Setenv("XDG_DATA_HOME", t.TempDir())
+	store := sessions.NewStore(sessions.StoreOptions{})
+	if _, err := store.Create(sessions.CreateInput{SessionID: "side", SessionKind: sessions.SessionKindSide}); err != nil {
+		t.Fatalf("create side session: %v", err)
+	}
+
+	err := preflightExecSession(execOptions{resumeLatest: true})
+	if err == nil || !strings.Contains(err.Error(), "No Zero sessions available to resume") {
+		t.Fatalf("preflightExecSession error = %v, want no resumable sessions", err)
+	}
+
+	err = preflightExecSession(execOptions{resume: "side"})
+	if err == nil || !strings.Contains(err.Error(), "Zero session is not resumable: side") {
+		t.Fatalf("explicit side-session preflight error = %v, want non-resumable rejection", err)
+	}
+}

internal/cli/exec_spec.go

@@ -46,6 +46,11 @@ type execSpecDraftRun struct {
 	reasoningEffort    string
 	specPermissionMode agent.PermissionMode
 	notifier           *notify.Notifier
+	// profilePolicy carries the selected execution profile's escalation policy
+	// (nil when none): the profile displaces resolved.MaxTurns before this
+	// path branches off, so the spec-draft run needs the same safety net as
+	// the main run.
+	profilePolicy *agent.ProfilePolicy
 }
 
 type execSpecDraftInfo struct {
@@ -119,6 +124,7 @@ func runExecSpecDraft(run execSpecDraftRun) int {
 		ProviderName:    run.resolved.Provider.Name,
 		Model:           run.resolved.Provider.Model,
 		ReasoningEffort: run.reasoningEffort,
+		Profile:         run.profilePolicy,
 		Cwd:             run.workspaceRoot,
 		SystemPrompt:    specmode.DraftSystemPrompt,
 		Images:          run.images,

internal/cli/exec_test.go

@@ -549,10 +549,14 @@ func TestExecMemberAutoToolListIncludesMutators(t *testing.T) {
 }
 
 func TestRunExecAcceptsLegacyModelProfileFlags(t *testing.T) {
+	// --profile (legacy, inert model profile) and --exec-profile (execution
+	// profile) are distinct flags and must coexist on one invocation.
 	exitCode, stdout, stderr := runExecWithEcho(t, []string{
 		"exec",
 		"--profile",
 		"fast",
+		"--exec-profile",
+		"balanced",
 		"--reasoning-effort",
 		"low",
 		"hello",

internal/cli/exec_turn_session_test.go

@@ -0,0 +1,125 @@
+package cli
+
+import (
+	"bytes"
+	"net/http"
+	"net/http/httptest"
+	"strings"
+	"sync/atomic"
+	"testing"
+	"time"
+
+	"github.com/Gitlawb/zero/internal/config"
+	"github.com/Gitlawb/zero/internal/providers/openai"
+	"github.com/Gitlawb/zero/internal/zeroruntime"
+)
+
+// TestRunExecOptimizedSessionUnderGate proves the end-to-end wiring: with
+// ZERO_OPENAI_TURN_SESSION on and a real official-OpenAI provider, a headless
+// run streams through the optimized session — the server sees the prewarm HEAD
+// probe in addition to the turn's POST.
+func TestRunExecOptimizedSessionUnderGate(t *testing.T) {
+	t.Setenv("XDG_DATA_HOME", t.TempDir())
+	t.Setenv("ZERO_OPENAI_TURN_SESSION", "1")
+	cwd := t.TempDir()
+
+	var heads, posts atomic.Int64
+	server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		switch {
+		case r.Method == http.MethodHead:
+			heads.Add(1)
+			w.WriteHeader(http.StatusMethodNotAllowed)
+		case r.Method == http.MethodPost && strings.HasSuffix(r.URL.Path, "/chat/completions"):
+			posts.Add(1)
+			w.Header().Set("Content-Type", "text/event-stream")
+			_, _ = w.Write([]byte("data: {\"choices\":[{\"delta\":{\"content\":\"done\"}}]}\n\n"))
+			_, _ = w.Write([]byte("data: [DONE]\n\n"))
+		default:
+			w.WriteHeader(http.StatusNotFound)
+		}
+	}))
+	defer server.Close()
+
+	var stdout, stderr bytes.Buffer
+	exitCode := runWithDeps([]string{"exec", "say done"}, &stdout, &stderr, appDeps{
+		getwd: func() (string, error) { return cwd, nil },
+		resolveConfig: func(string, config.Overrides) (config.ResolvedConfig, error) {
+			return config.ResolvedConfig{
+				ActiveProvider: "official-openai",
+				Provider: config.ProviderProfile{
+					Name:         "official-openai",
+					ProviderKind: config.ProviderKindOpenAI,
+					BaseURL:      server.URL,
+					APIKey:       "sk-exec-test",
+					Model:        "pr8-exec-model",
+				},
+			}, nil
+		},
+		newProvider: func(profile config.ProviderProfile) (zeroruntime.Provider, error) {
+			// Pin a keep-alive transport so the probe fires on every platform:
+			// the shared default transport disables keep-alives on macOS, where
+			// the session correctly skips the prewarm.
+			transport := http.DefaultTransport.(*http.Transport).Clone()
+			transport.DisableKeepAlives = false
+			return openai.New(openai.Options{
+				APIKey:     profile.APIKey,
+				BaseURL:    profile.BaseURL,
+				Model:      profile.Model,
+				HTTPClient: &http.Client{Transport: transport},
+			})
+		},
+	})
+	if exitCode != exitSuccess {
+		t.Fatalf("exitCode = %d stdout=%s stderr=%s", exitCode, stdout.String(), stderr.String())
+	}
+	if got := posts.Load(); got < 1 {
+		t.Fatalf("server saw %d chat-completions POSTs, want >= 1", got)
+	}
+	// The prewarm probe is asynchronous; it fires well before the model turn
+	// finishes, but poll rather than assuming scheduling order. The deadline
+	// exceeds the production prewarmTimeout (3s) so a slow probe cannot flake
+	// this assertion.
+	deadline := time.Now().Add(5 * time.Second)
+	for heads.Load() == 0 && time.Now().Before(deadline) {
+		time.Sleep(10 * time.Millisecond)
+	}
+	if got := heads.Load(); got != 1 {
+		t.Fatalf("server saw %d prewarm HEAD probes, want exactly 1", got)
+	}
+}
+
+// TestRunExecGateOnFallsBackForFakeProvider proves fallback safety: with the
+// gate on but a provider that is not the concrete *openai.Provider, the run
+// proceeds on the default path exactly as today.
+func TestRunExecGateOnFallsBackForFakeProvider(t *testing.T) {
+	t.Setenv("XDG_DATA_HOME", t.TempDir())
+	t.Setenv("ZERO_OPENAI_TURN_SESSION", "1")
+	cwd := t.TempDir()
+
+	var builds int
+	var stdout, stderr bytes.Buffer
+	exitCode := runWithDeps([]string{"exec", "--model", "claude-haiku-4.5", "hello"}, &stdout, &stderr, appDeps{
+		getwd: func() (string, error) { return cwd, nil },
+		resolveConfig: func(_ string, overrides config.Overrides) (config.ResolvedConfig, error) {
+			model := "claude-haiku-4.5"
+			if overrides.Provider.Model != "" {
+				model = overrides.Provider.Model
+			}
+			cfg := execResolvedConfig()
+			cfg.Provider.ProviderKind = config.ProviderKindAnthropic
+			cfg.Provider.Model = model
+			cfg.MaxTurns = 3
+			return cfg, nil
+		},
+		newProvider: func(config.ProviderProfile) (zeroruntime.Provider, error) {
+			builds++
+			return &escalatingExecProvider{}, nil
+		},
+	})
+	if exitCode != exitSuccess {
+		t.Fatalf("exitCode = %d stdout=%s stderr=%s", exitCode, stdout.String(), stderr.String())
+	}
+	if builds != 1 {
+		t.Fatalf("newProvider called %d times, want 1 (default path, no optimized session)", builds)
+	}
+}

internal/cli/provider_onboarding.go

@@ -53,6 +53,22 @@ func runProvidersUse(args []string, stdout io.Writer, stderr io.Writer, deps app
 	if err != nil {
 		return writeAppError(stderr, err.Error(), exitCrash)
 	}
+	// SetActiveProvider only ever matches profiles persisted in config.json
+	// (see config.ProviderPersisted), but a provider can be visible in
+	// `zero providers list`/the TUI picker purely because Resolve()
+	// synthesized it in-memory from an ambient env var (e.g. OPENAI_API_KEY)
+	// without ever writing a row to disk. Without this check, switching to
+	// that provider by name always fails with a confusing "not found" even
+	// though it is genuinely usable this session (issue #707).
+	persisted, err := config.ProviderPersisted(configPath, options.name)
+	if err != nil {
+		return writeAppError(stderr, err.Error(), exitCrash)
+	}
+	if !persisted {
+		if exit, handled := reportUnpersistedProviderUse(stdout, stderr, deps, options, configPath); handled {
+			return exit
+		}
+	}
 	cfg, err := config.SetActiveProvider(configPath, options.name)
 	if err != nil {
 		return writeAppError(stderr, err.Error(), exitCrash)
@@ -352,6 +368,22 @@ func runProvidersRemove(args []string, stdout io.Writer, stderr io.Writer, deps
 		return writeAppError(stderr, err.Error(), exitCrash)
 	}
 	name := options.names[0]
+	// RemoveProvider only ever matches profiles persisted in config.json (see
+	// config.ProviderPersisted), but a provider can be visible in
+	// `zero providers list`/the TUI picker purely because Resolve()
+	// synthesized it in-memory from an ambient env var (e.g. OPENAI_API_KEY)
+	// without ever writing a row to disk. Without this check, deleting that
+	// provider by name always fails with a confusing "not found" even though
+	// it is genuinely visible/usable this session (issue #707).
+	persisted, err := config.ProviderPersisted(configPath, name)
+	if err != nil {
+		return writeAppError(stderr, err.Error(), exitCrash)
+	}
+	if !persisted {
+		if exit, handled := reportUnpersistedProviderRemove(stdout, stderr, deps, name, options.json, configPath); handled {
+			return exit
+		}
+	}
 	cfg, err := config.RemoveProvider(configPath, name)
 	if err != nil {
 		return writeAppError(stderr, err.Error(), exitCrash)
@@ -428,6 +460,16 @@ func runProvidersRename(args []string, stdout io.Writer, stderr io.Writer, deps
 	if err != nil {
 		return writeAppError(stderr, err.Error(), exitCrash)
 	}
+	oldName := options.names[0]
+	persisted, err := config.ProviderPersisted(configPath, oldName)
+	if err != nil {
+		return writeAppError(stderr, err.Error(), exitCrash)
+	}
+	if !persisted {
+		if exit, handled := reportUnpersistedProviderRename(stdout, stderr, deps, oldName, options.json, configPath); handled {
+			return exit
+		}
+	}
 	cfg, err := config.RenameProvider(configPath, options.names[0], options.names[1])
 	if err != nil {
 		return writeAppError(stderr, err.Error(), exitCrash)
@@ -447,3 +489,120 @@ func runProvidersRename(args []string, stdout io.Writer, stderr io.Writer, deps
 	}
 	return exitSuccess
 }
+
+// providerResolvedByName reports whether name matches a provider in a
+// resolved provider list — used to tell a genuinely unknown name (a typo)
+// apart from a real, env-derived provider that just has no config.json row.
+func providerResolvedByName(providers []config.ProviderProfile, name string) bool {
+	name = strings.TrimSpace(name)
+	for _, provider := range providers {
+		if strings.EqualFold(strings.TrimSpace(provider.Name), name) {
+			return true
+		}
+	}
+	return false
+}
+
+// reportUnpersistedProviderUse handles `zero providers use <name>` for a
+// provider that is not persisted in config.json. If it's not resolvable at
+// all (an unknown/misspelled name), it returns handled=false so the caller
+// falls through to SetActiveProvider's real "not found" error. If it IS
+// resolvable — an env-derived profile (e.g. ambient OPENAI_API_KEY) —
+// SetActiveProvider would only ever fail "not found" against it, so this
+// reports the situation plainly instead of that confusing error (issue
+// #707).
+func reportUnpersistedProviderUse(stdout, stderr io.Writer, deps appDeps, options providerUseOptions, configPath string) (int, bool) {
+	resolved, exitCode := resolveCommandCenterConfig(stderr, deps)
+	if exitCode != exitSuccess {
+		// resolveCommandCenterConfig already wrote its own error to stderr;
+		// stop here instead of letting the caller try SetActiveProvider too.
+		return exitCode, true
+	}
+	if !providerResolvedByName(resolved.Providers, options.name) {
+		return exitCode, false
+	}
+	message := fmt.Sprintf(
+		"Provider %q is not saved in config.json (likely set via an environment variable), so there is no saved profile to switch to.\nIt is available whenever its environment variable is set, but is only active when selected (for example via ZERO_PROVIDER); unset its environment variable to stop Zero from detecting it automatically.",
+		options.name,
+	)
+	if options.json {
+		if err := writePrettyJSON(stdout, map[string]any{
+			"activeProvider": resolved.ActiveProvider,
+			"configPath":     configPath,
+			"persisted":      false,
+			"message":        message,
+		}); err != nil {
+			return exitCrash, true
+		}
+		return exitSuccess, true
+	}
+	if _, err := fmt.Fprintln(stdout, message); err != nil {
+		return exitCrash, true
+	}
+	return exitSuccess, true
+}
+
+// reportUnpersistedProviderRemove handles `zero providers remove <name>` for
+// a provider that is not persisted in config.json, mirroring the TUI
+// provider manager's delete handling (internal/tui/provider_manager.go). If
+// name isn't resolvable at all, it returns handled=false so the caller falls
+// through to RemoveProvider's real "not found" error.
+func reportUnpersistedProviderRemove(stdout, stderr io.Writer, deps appDeps, name string, jsonOutput bool, configPath string) (int, bool) {
+	resolved, exitCode := resolveCommandCenterConfig(stderr, deps)
+	if exitCode != exitSuccess {
+		return exitCode, true
+	}
+	if !providerResolvedByName(resolved.Providers, name) {
+		return exitCode, false
+	}
+	message := fmt.Sprintf(
+		"Provider %q is not saved in config.json (likely set via an environment variable) — nothing to remove there.\nUnset its environment variable to stop Zero from detecting it automatically.",
+		name,
+	)
+	if jsonOutput {
+		if err := writePrettyJSON(stdout, map[string]any{
+			"removed":        "",
+			"keyRemoved":     false,
+			"activeProvider": resolved.ActiveProvider,
+			"configPath":     configPath,
+			"persisted":      false,
+			"message":        message,
+		}); err != nil {
+			return exitCrash, true
+		}
+		return exitSuccess, true
+	}
+	if _, err := fmt.Fprintln(stdout, message); err != nil {
+		return exitCrash, true
+	}
+	return exitSuccess, true
+}
+
+func reportUnpersistedProviderRename(stdout, stderr io.Writer, deps appDeps, name string, jsonOutput bool, configPath string) (int, bool) {
+	resolved, exitCode := resolveCommandCenterConfig(stderr, deps)
+	if exitCode != exitSuccess {
+		return exitCode, true
+	}
+	if !providerResolvedByName(resolved.Providers, name) {
+		return exitCode, false
+	}
+	message := fmt.Sprintf(
+		"Provider %q is not saved in config.json (likely set via an environment variable), so there is no saved profile to rename.",
+		name,
+	)
+	if jsonOutput {
+		if err := writePrettyJSON(stdout, map[string]any{
+			"renamed":    nil,
+			"configPath": configPath,
+			"persisted":  false,
+			"message":    message,
+		}); err != nil {
+			return exitCrash, true
+		}
+		return exitSuccess, true
+	}
+	if _, err := fmt.Fprintln(stdout, message); err != nil {
+		return exitCrash, true
+	}
+	return exitSuccess, true
+}

internal/cli/provider_onboarding_test.go

@@ -75,6 +75,111 @@ func TestRunProvidersUseJSONIncludesActiveProviderAndConfigPath(t *testing.T) {
 	}
 }
 
+func TestRunProvidersUseSurfacesMalformedConfig(t *testing.T) {
+	var stdout, stderr bytes.Buffer
+	configPath := filepath.Join(t.TempDir(), "config.json")
+	if err := os.WriteFile(configPath, []byte("{"), 0o600); err != nil {
+		t.Fatalf("write malformed config: %v", err)
+	}
+
+	exitCode := runWithDeps([]string{"providers", "use", "openai"}, &stdout, &stderr, providerSetupDeps(configPath))
+
+	if exitCode != exitCrash {
+		t.Fatalf("exit code = %d, want %d", exitCode, exitCrash)
+	}
+	if !strings.Contains(stderr.String(), "invalid config JSON") {
+		t.Fatalf("stderr = %q, want malformed-config error", stderr.String())
+	}
+}
+
+func TestRunProvidersUseEnvDerivedJSONIncludesConfigPath(t *testing.T) {
+	t.Setenv("OPENAI_API_KEY", "sk-env")
+	var stdout, stderr bytes.Buffer
+	configPath := filepath.Join(t.TempDir(), "config.json")
+	writeProviderOnboardingConfig(t, configPath, config.FileConfig{})
+
+	exitCode := runWithDeps([]string{"providers", "use", "openai", "--json"}, &stdout, &stderr, providerSetupDeps(configPath))
+
+	if exitCode != exitSuccess {
+		t.Fatalf("exit code = %d, want %d: %s", exitCode, exitSuccess, stderr.String())
+	}
+	var payload struct {
+		ConfigPath string `json:"configPath"`
+		Persisted  bool   `json:"persisted"`
+	}
+	if err := json.Unmarshal(stdout.Bytes(), &payload); err != nil {
+		t.Fatalf("decode JSON: %v\n%s", err, stdout.String())
+	}
+	if payload.ConfigPath != configPath || payload.Persisted {
+		t.Fatalf("unexpected payload: %+v", payload)
+	}
+}
+
+func TestRunProvidersRemoveEnvDerivedJSONKeepsSchema(t *testing.T) {
+	t.Setenv("OPENAI_API_KEY", "sk-env")
+	var stdout, stderr bytes.Buffer
+	configPath := filepath.Join(t.TempDir(), "config.json")
+	writeProviderOnboardingConfig(t, configPath, config.FileConfig{})
+
+	exitCode := runWithDeps([]string{"providers", "remove", "openai", "--json"}, &stdout, &stderr, providerSetupDeps(configPath))
+
+	if exitCode != exitSuccess {
+		t.Fatalf("exit code = %d, want %d: %s", exitCode, exitSuccess, stderr.String())
+	}
+	var payload struct {
+		Removed    string `json:"removed"`
+		KeyRemoved bool   `json:"keyRemoved"`
+		Persisted  bool   `json:"persisted"`
+		ConfigPath string `json:"configPath"`
+	}
+	if err := json.Unmarshal(stdout.Bytes(), &payload); err != nil {
+		t.Fatalf("decode JSON: %v\n%s", err, stdout.String())
+	}
+	if payload.Removed != "" || payload.KeyRemoved || payload.Persisted || payload.ConfigPath != configPath {
+		t.Fatalf("unexpected payload: %+v", payload)
+	}
+}
+
+func TestRunProvidersRenameEnvDerivedExplainsNoSavedProfile(t *testing.T) {
+	t.Setenv("OPENAI_API_KEY", "sk-env")
+	var stdout, stderr bytes.Buffer
+	configPath := filepath.Join(t.TempDir(), "config.json")
+	writeProviderOnboardingConfig(t, configPath, config.FileConfig{})
+
+	exitCode := runWithDeps([]string{"providers", "rename", "openai", "renamed"}, &stdout, &stderr, providerSetupDeps(configPath))
+
+	if exitCode != exitSuccess {
+		t.Fatalf("exit code = %d, want %d: %s", exitCode, exitSuccess, stderr.String())
+	}
+	if !strings.Contains(stdout.String(), "no saved profile to rename") {
+		t.Fatalf("stdout = %q, want unpersisted explanation", stdout.String())
+	}
+}
+
+func TestRunProvidersRenameEnvDerivedJSONKeepsSchema(t *testing.T) {
+	t.Setenv("OPENAI_API_KEY", "sk-env")
+	var stdout, stderr bytes.Buffer
+	configPath := filepath.Join(t.TempDir(), "config.json")
+	writeProviderOnboardingConfig(t, configPath, config.FileConfig{})
+
+	exitCode := runWithDeps([]string{"providers", "rename", "openai", "renamed", "--json"}, &stdout, &stderr, providerSetupDeps(configPath))
+
+	if exitCode != exitSuccess {
+		t.Fatalf("exit code = %d, want %d: %s", exitCode, exitSuccess, stderr.String())
+	}
+	var payload struct {
+		Renamed    *struct{} `json:"renamed"`
+		ConfigPath string    `json:"configPath"`
+		Persisted  bool      `json:"persisted"`
+	}
+	if err := json.Unmarshal(stdout.Bytes(), &payload); err != nil {
+		t.Fatalf("decode JSON: %v\n%s", err, stdout.String())
+	}
+	if payload.Renamed != nil || payload.ConfigPath != configPath || payload.Persisted {
+		t.Fatalf("unexpected payload: %+v", payload)
+	}
+}
+
 func TestRunProvidersUseRejectsUsageErrors(t *testing.T) {
 	cases := []struct {
 		name string

internal/cli/sandbox_test.go

@@ -1001,3 +1001,18 @@ func sandboxPolicyCapabilityStatus(capabilities []sandbox.BackendCapability, key
 	}
 	return ""
 }
+
+func TestApplyConfiguredSandboxPolicyEnabledFalseDisables(t *testing.T) {
+	disabled := false
+	if got := applyConfiguredSandboxPolicy(sandbox.DefaultPolicy(), config.SandboxConfig{Enabled: &disabled}); got.Mode != sandbox.ModeDisabled {
+		t.Fatalf("enabled:false must yield ModeDisabled, got %q", got.Mode)
+	}
+	// Omitted (nil) and explicit true keep the default enforcing mode.
+	if got := applyConfiguredSandboxPolicy(sandbox.DefaultPolicy(), config.SandboxConfig{}); got.Mode == sandbox.ModeDisabled {
+		t.Fatal("omitted enabled must not disable the sandbox")
+	}
+	enabled := true
+	if got := applyConfiguredSandboxPolicy(sandbox.DefaultPolicy(), config.SandboxConfig{Enabled: &enabled}); got.Mode == sandbox.ModeDisabled {
+		t.Fatal("enabled:true must not disable the sandbox")
+	}
+}

internal/cli/sessions.go

@@ -216,10 +216,10 @@ func parseNonEmptySessionsFlag(flag string, value string) (string, error) {
 func parseSessionKindFlag(value string) (sessions.SessionKind, error) {
 	kind := sessions.SessionKind(strings.ToLower(strings.TrimSpace(value)))
 	switch kind {
-	case sessions.SessionKindFork, sessions.SessionKindChild, sessions.SessionKindSpecDraft, sessions.SessionKindSpecImpl:
+	case sessions.SessionKindFork, sessions.SessionKindChild, sessions.SessionKindSide, sessions.SessionKindSpecDraft, sessions.SessionKindSpecImpl:
 		return kind, nil
 	default:
-		return "", execUsageError{fmt.Sprintf("invalid --kind %q. Expected fork, child, spec-draft, or spec-impl.", value)}
+		return "", execUsageError{fmt.Sprintf("invalid --kind %q. Expected fork, child, side, spec-draft, or spec-impl.", value)}
 	}
 }
 
@@ -535,7 +535,7 @@ Commands:
 
 Flags:
       --json            Print JSON output
-      --kind <kind>     Filter list by fork, child, spec-draft, or spec-impl
+      --kind <kind>     Filter list by fork, child, side, spec-draft, or spec-impl
       --sequence <n>    Rewind target sequence (rewind-plan, rewind)
       --event <id>      Rewind target event id (rewind-plan, rewind)
       --exclude-target  Drop the target event (rewind-plan, rewind)

internal/cli/sessions_test.go

@@ -233,6 +233,9 @@ func TestRunSessionsListFiltersByKind(t *testing.T) {
 	}); err != nil {
 		t.Fatal(err)
 	}
+	if _, err := store.Create(sessions.CreateInput{SessionID: "side", SessionKind: sessions.SessionKindSide, Title: "BTW side"}); err != nil {
+		t.Fatal(err)
+	}
 
 	var stdout bytes.Buffer
 	var stderr bytes.Buffer
@@ -249,6 +252,17 @@ func TestRunSessionsListFiltersByKind(t *testing.T) {
 		t.Fatalf("filtered sessions output = %q", output)
 	}
 
+	stdout.Reset()
+	stderr.Reset()
+	exitCode = runWithDeps([]string{"sessions", "list", "--kind", "side"}, &stdout, &stderr, appDeps{
+		newSessionStore: func() *sessions.Store {
+			return store
+		},
+	})
+	if exitCode != exitSuccess || !strings.Contains(stdout.String(), "side") || strings.Contains(stdout.String(), "regular") {
+		t.Fatalf("sessions list --kind side exit=%d stdout=%q stderr=%q", exitCode, stdout.String(), stderr.String())
+	}
+
 	stdout.Reset()
 	stderr.Reset()
 	exitCode = runWithDeps([]string{"sessions", "tree", "draft", "--kind", "spec-draft"}, &stdout, &stderr, appDeps{

internal/config/resolver.go

@@ -93,6 +93,16 @@ func Resolve(options ResolveOptions) (ResolvedConfig, error) {
 		if err != nil {
 			return ResolvedConfig{}, err
 		}
+		// Sandbox.Enabled is NOT accepted from a provider command, for the same
+		// reason project config cannot set it (see mergeProjectConfig): only
+		// global config and the CLI may turn the sandbox off. A provider command
+		// is an arbitrary executable named in config, and LoadProviderCommand
+		// parses its stdout into a full FileConfig — so without this, a command
+		// returning a valid provider plus {"sandbox":{"enabled":false}} would
+		// disable the very sandbox meant to constrain what it can do. Cleared
+		// here rather than in mergeConfig because that helper is shared with the
+		// trusted global-config merge, which must keep honouring the setting.
+		commandConfig.Sandbox.Enabled = nil
 		mergeConfig(&cfg, commandConfig)
 	}
 
@@ -216,6 +226,9 @@ func mergeConfig(dst *FileConfig, src FileConfig) {
 		mergeProvider(dst, provider)
 	}
 	mergeMCPConfig(&dst.MCP, src.MCP, true)
+	if src.Sandbox.Enabled != nil {
+		dst.Sandbox.Enabled = src.Sandbox.Enabled
+	}
 	if network := strings.TrimSpace(src.Sandbox.Network); network != "" {
 		dst.Sandbox.Network = network
 	}
@@ -273,6 +286,10 @@ func mergeProjectConfig(dst *FileConfig, src FileConfig) error {
 	if err := mergeProjectMCPConfig(&dst.MCP, src.MCP); err != nil {
 		return err
 	}
+	// Sandbox.Enabled is intentionally NOT merged from project config: a cloned
+	// repo's .zero/config.json must not be able to disable the sandbox that
+	// constrains it. Only global config and CLI can turn the sandbox off.
+	//
 	// Sandbox.AdditionalWriteRoots is intentionally NOT merged from project
 	// config: a cloned repo's .zero/config.json must not be able to grant
 	// itself write access outside the workspace. Global config and CLI flags
@@ -688,6 +705,9 @@ func applyOverrides(cfg *FileConfig, overrides Overrides) {
 	if overrides.MaxTurns > 0 {
 		cfg.MaxTurns = overrides.MaxTurns
 	}
+	if overrides.Sandbox.Enabled != nil {
+		cfg.Sandbox.Enabled = overrides.Sandbox.Enabled
+	}
 	if overrides.Sandbox.BlockUnixSockets {
 		cfg.Sandbox.BlockUnixSockets = true
 	}

internal/config/resolver_test.go

@@ -1744,6 +1744,85 @@ func TestResolveSandboxNetworkProjectConfigCannotWeaken(t *testing.T) {
 	}
 }
 
+func TestResolveSandboxEnabledUserConfigDisables(t *testing.T) {
+	userPath := writeConfig(t, `{"sandbox": {"enabled": false}}`)
+	resolved, err := Resolve(ResolveOptions{UserConfigPath: userPath, Env: map[string]string{}})
+	if err != nil {
+		t.Fatalf("Resolve: %v", err)
+	}
+	if resolved.Sandbox.Enabled == nil || *resolved.Sandbox.Enabled {
+		t.Fatalf("user config enabled:false must resolve to a false pointer, got %v", resolved.Sandbox.Enabled)
+	}
+}
+
+func TestResolveSandboxDisableIgnoredFromProjectConfig(t *testing.T) {
+	// Security: a cloned repo's project config must NOT be able to disable the
+	// sandbox that constrains it. Only global config / CLI can turn it off.
+	userPath := writeConfig(t, `{}`)
+	projectPath := writeConfig(t, `{"sandbox": {"enabled": false}}`)
+	resolved, err := Resolve(ResolveOptions{UserConfigPath: userPath, ProjectConfigPath: projectPath, Env: map[string]string{}})
+	if err != nil {
+		t.Fatalf("Resolve: %v", err)
+	}
+	if resolved.Sandbox.Enabled != nil {
+		t.Fatalf("project config enabled:false must be ignored, got %v — a repo could disable the sandbox", *resolved.Sandbox.Enabled)
+	}
+}
+
+func TestResolveSandboxEnabledIgnoredFromProviderCommand(t *testing.T) {
+	// Security: a provider command is an arbitrary executable named in config, and
+	// its stdout is parsed into a full FileConfig — so it must NOT be able to reach
+	// Sandbox.Enabled, in either direction. Same rule as project config: only global
+	// config / CLI may turn the sandbox off.
+	for _, tc := range []struct {
+		name  string
+		value string
+	}{
+		{name: "cannot disable", value: "false"},
+		{name: "cannot enable", value: "true"},
+	} {
+		t.Run(tc.name, func(t *testing.T) {
+			command := writeCommand(t, commandScript{
+				Stdout: `{"activeProvider":"cmd","providers":[{"name":"cmd","provider":"openai","apiKey":"sk-command","model":"gpt-command"}],"sandbox":{"enabled":` + tc.value + `}}`,
+			})
+			resolved, err := Resolve(ResolveOptions{ProviderCommand: command, Env: map[string]string{}})
+			if err != nil {
+				t.Fatalf("Resolve: %v", err)
+			}
+			// The provider itself must still be applied — only the sandbox toggle is dropped.
+			if resolved.ActiveProvider != "cmd" {
+				t.Fatalf("ActiveProvider = %q, want cmd (provider command must still apply)", resolved.ActiveProvider)
+			}
+			if resolved.Sandbox.Enabled != nil {
+				t.Fatalf("provider command set sandbox.enabled=%s, got %v — a provider command could toggle the sandbox", tc.value, *resolved.Sandbox.Enabled)
+			}
+		})
+	}
+}
+
+func TestResolveSandboxEnabledCLIOverride(t *testing.T) {
+	// A CLI/programmatic override (applied after every config source) must be able
+	// to disable the sandbox, even with no config file.
+	disabled := false
+	resolved, err := Resolve(ResolveOptions{Env: map[string]string{}, Overrides: Overrides{Sandbox: SandboxConfig{Enabled: &disabled}}})
+	if err != nil {
+		t.Fatalf("Resolve: %v", err)
+	}
+	if resolved.Sandbox.Enabled == nil || *resolved.Sandbox.Enabled {
+		t.Fatalf("CLI override enabled:false must disable the sandbox, got %v", resolved.Sandbox.Enabled)
+	}
+	// An explicit true override must win over a lower-precedence global-config false.
+	userPath := writeConfig(t, `{"sandbox": {"enabled": false}}`)
+	enabled := true
+	resolved2, err := Resolve(ResolveOptions{UserConfigPath: userPath, Env: map[string]string{}, Overrides: Overrides{Sandbox: SandboxConfig{Enabled: &enabled}}})
+	if err != nil {
+		t.Fatalf("Resolve: %v", err)
+	}
+	if resolved2.Sandbox.Enabled == nil || !*resolved2.Sandbox.Enabled {
+		t.Fatalf("CLI override enabled:true must win over config false, got %v", resolved2.Sandbox.Enabled)
+	}
+}
+
 func TestResolveNotifyValid(t *testing.T) {
 	path := writeConfig(t, `{"notify":{"mode":"both","focusMode":"always"}}`)
 	resolved, err := Resolve(ResolveOptions{UserConfigPath: path, Env: map[string]string{}})

internal/config/types.go

@@ -62,6 +62,12 @@ func HasProviderProfile(profile ProviderProfile) bool {
 }
 
 type SandboxConfig struct {
+	// Enabled turns the sandbox off when set to false (`"sandbox": {"enabled":
+	// false}`). A pointer distinguishes an explicit false (disable) from an
+	// omitted key (keep the default: enabled). Honored from the GLOBAL user
+	// config and CLI only — deliberately NOT project config, so a cloned repo
+	// cannot disable the sandbox that constrains it. nil / true keep enforcement.
+	Enabled *bool `json:"enabled,omitempty"`
 	// Network controls whether shell commands classified as network-touching
 	// (curl, git push, package installs, …) are allowed: "allow" or "deny".
 	// Empty keeps the built-in default (deny). Without this knob the engine's

internal/config/writer.go

@@ -177,6 +177,33 @@ func SetActiveProvider(path string, name string) (FileConfig, error) {
 	return FileConfig{}, fmt.Errorf("provider %q not found", name)
 }
 
+// ProviderPersisted reports whether a provider profile named name actually has
+// a row in the config file at path. A provider can appear in the resolved/
+// in-memory provider list without ever being written to config.json — e.g.
+// applyProviderEnv synthesizes an "openai" profile purely from an ambient
+// OPENAI_API_KEY environment variable on every Resolve() call, without ever
+// persisting it. RemoveProvider/SetActiveProvider/SetProviderModel only ever
+// look at what's on disk, so a caller offering to mutate a provider by name
+// should check this first: "not on disk" needs different handling (nothing to
+// persist/remove there) than a name that doesn't exist anywhere at all.
+func ProviderPersisted(path string, name string) (bool, error) {
+	path = strings.TrimSpace(path)
+	name = strings.TrimSpace(name)
+	if path == "" || name == "" {
+		return false, nil
+	}
+	cfg, err := loadConfigFile(path)
+	if err != nil {
+		return false, err
+	}
+	for _, provider := range cfg.Providers {
+		if strings.EqualFold(strings.TrimSpace(provider.Name), name) {
+			return true, nil
+		}
+	}
+	return false, nil
+}
+
 // RemoveProvider deletes the named provider profile from the config at path.
 // When the removed profile was active, activeProvider hands off to the first
 // remaining provider (or clears when none remain) so the config never points at

internal/execprofile/profile.go

@@ -0,0 +1,153 @@
+// Package execprofile defines named execution profiles: bundles of loop-posture
+// knobs (turn budget, reasoning effort, self-correction, escalation triggers)
+// that tune how hard the agent loop tries, independent of which model runs.
+// Profiles COMPOSE with --mode: a mode answers "what runs" (model, tools),
+// a profile answers "how hard the loop tries". Selection precedence is
+// explicit flag > --mode > profile, enforced by the callers applying profiles
+// last with the same fill-only-if-unset rule modes use.
+//
+// The catalog is deliberately tiny and value-stable: balanced is the empty
+// profile (selecting it leaves the run's options, budget, and loop behavior
+// identical to an unflagged run — asserted by test; only the opt-in trace and
+// bench artifacts record the selected profile name, by design, so captures are
+// self-describing), fast trades turn budget and effort for latency with a
+// one-shot escalation back to the displaced posture as the safety net, and
+// thorough raises the budget and arms full self-correction.
+package execprofile
+
+import (
+	"sort"
+	"strings"
+
+	"github.com/Gitlawb/zero/internal/agent"
+	"github.com/Gitlawb/zero/internal/sandbox"
+)
+
+// Profile is a named execution posture. Zero-valued fields inherit the run's
+// existing value (flag, mode, config, or built-in default) — a profile only
+// ever fills knobs the caller left unset, except MaxTurns which REPLACES the
+// resolved budget (that displacement is what escalation restores).
+type Profile struct {
+	// Name identifies the profile in traces, help text, and diagnostics.
+	Name string
+	// MaxTurns replaces the resolved per-run tool-turn budget when > 0 and the
+	// user did not pass an explicit --max-turns. The displaced resolved value
+	// becomes the escalation target (see Policy).
+	MaxTurns int
+	// ReasoningEffort fills the run's effort when the user left it unset. It
+	// still flows through the model-supported gating at the call site, so an
+	// unsupported level degrades exactly like an explicit flag would.
+	ReasoningEffort string
+	// SelfCorrect arms the post-edit verify-and-correct loop. Profiles can only
+	// turn it ON (the flag is presence-only, so false is indistinguishable from
+	// unset and must never override an explicit opt-in).
+	SelfCorrect bool
+
+	// Escalation triggers. All zero means the profile never escalates and
+	// Policy returns nil (the loop stays byte-identical to a nil-profile run).
+	// Targets are NOT part of the catalog: escalation restores the values the
+	// profile displaced at selection time, never invented ones, so targets are
+	// stamped by Policy from what the caller measured.
+	EscalateOnToolFailureStreak   int
+	EscalateOnCompletionUncertain int
+	EscalateOnSelfCorrectFailure  bool
+	EscalateOnRiskyMutation       sandbox.RiskLevel
+}
+
+// The catalog. Values marked provisional are floors/ceilings derived from the
+// Phase 0 baseline's read-class evidence (successful nav runs used single-digit
+// turns; the mutating classes produced no successful samples to tune from) and
+// are expected to be re-tuned from the post-oracle-fix re-capture. The
+// escalation safety net is what makes shipping provisional Fast values safe:
+// a run that hits trouble gets its displaced budget back mid-run.
+var (
+	// Balanced is the default posture: empty on purpose. Selecting it changes
+	// nothing — the no-regression invariant of the whole feature.
+	Balanced = Profile{Name: "balanced"}
+
+	// Fast starts cheap (30 turns, low effort) and arms every escalation
+	// trigger so a struggling run restores its displaced posture: two
+	// same-tool retriable failures in a row, a second uncertain completion
+	// evaluation, a failing self-correction cycle, or a critical-risk
+	// mutation. The risk threshold is deliberately Critical, not High: the
+	// sandbox classifies EVERY shell command as at least high risk before any
+	// command analysis, so a High trigger would fire on the first `go test`
+	// of virtually any coding task and spend the one-shot on turn one.
+	// Critical marks the genuinely scary categories (destructive commands,
+	// piped installers, out-of-workspace writes, network mutations).
+	Fast = Profile{
+		Name:                          "fast",
+		MaxTurns:                      30,
+		ReasoningEffort:               "low",
+		EscalateOnToolFailureStreak:   2,
+		EscalateOnCompletionUncertain: 2,
+		EscalateOnSelfCorrectFailure:  true,
+		EscalateOnRiskyMutation:       sandbox.RiskCritical,
+	}
+
+	// Thorough doubles the default budget, asks for high effort, and arms the
+	// full post-edit verify-and-correct loop. Already the maximum posture, so
+	// it has nothing to escalate to.
+	Thorough = Profile{
+		Name:            "thorough",
+		MaxTurns:        160,
+		ReasoningEffort: "high",
+		SelfCorrect:     true,
+	}
+)
+
+var catalog = map[string]Profile{
+	Balanced.Name: Balanced,
+	Fast.Name:     Fast,
+	Thorough.Name: Thorough,
+}
+
+// Lookup resolves a profile by name, case-insensitively and ignoring
+// surrounding whitespace.
+func Lookup(name string) (Profile, bool) {
+	profile, ok := catalog[strings.ToLower(strings.TrimSpace(name))]
+	return profile, ok
+}
+
+// Names returns the catalog's profile names, sorted, for usage errors and help.
+func Names() []string {
+	names := make([]string, 0, len(catalog))
+	for name := range catalog {
+		names = append(names, name)
+	}
+	sort.Strings(names)
+	return names
+}
+
+// Policy projects the profile into the loop-facing ProfilePolicy.
+//
+// displacedMaxTurns is the turn budget this profile displaced at selection time
+// (what the run would have used without it) and becomes the escalation's
+// restore target; pass 0 when the profile did not displace the budget (e.g.
+// the user pinned --max-turns explicitly) so escalation leaves the ceiling
+// untouched. effortFilled reports whether the profile actually filled the
+// run's reasoning effort (it backs off when the user set one explicitly); the
+// displaced effort is then "" (the provider default) by construction, which
+// the escalation restores via RestoreDefaultEffort — a plain "" target would
+// mean "leave untouched".
+//
+// Profiles with no armed triggers return nil: the loop treats a nil policy as
+// "no profile" (no observation, no counters), which keeps balanced and
+// thorough runs byte-identical to the same knob values set by hand.
+func (p Profile) Policy(displacedMaxTurns int, effortFilled bool) *agent.ProfilePolicy {
+	if p.EscalateOnToolFailureStreak == 0 && p.EscalateOnCompletionUncertain == 0 &&
+		!p.EscalateOnSelfCorrectFailure && p.EscalateOnRiskyMutation == "" {
+		return nil
+	}
+	return &agent.ProfilePolicy{
+		Name: p.Name,
+		Escalate: &agent.PostureEscalation{
+			MaxTurns:              displacedMaxTurns,
+			RestoreDefaultEffort:  effortFilled,
+			OnToolFailureStreak:   p.EscalateOnToolFailureStreak,
+			OnCompletionUncertain: p.EscalateOnCompletionUncertain,
+			OnSelfCorrectFailure:  p.EscalateOnSelfCorrectFailure,
+			OnRiskyMutation:       p.EscalateOnRiskyMutation,
+		},
+	}
+}

internal/execprofile/profile_test.go

@@ -0,0 +1,131 @@
+package execprofile
+
+import (
+	"reflect"
+	"testing"
+
+	"github.com/Gitlawb/zero/internal/sandbox"
+)
+
+// The no-regression invariant of the whole feature: balanced must be the empty
+// profile (nothing but a name), so selecting it cannot change a run.
+func TestBalancedProfileIsEmpty(t *testing.T) {
+	profile, ok := Lookup("balanced")
+	if !ok {
+		t.Fatal("balanced must exist in the catalog")
+	}
+	if profile.Name != "balanced" {
+		t.Fatalf("Name = %q, want %q", profile.Name, "balanced")
+	}
+	profile.Name = ""
+	if !reflect.DeepEqual(profile, Profile{}) {
+		t.Fatalf("balanced must be zero-valued apart from its name, got %+v", profile)
+	}
+	if policy := Balanced.Policy(80, false); policy != nil {
+		t.Fatalf("balanced Policy must be nil (byte-identical loop), got %+v", policy)
+	}
+}
+
+func TestLookupIsCaseAndSpaceInsensitive(t *testing.T) {
+	for _, name := range []string{"fast", "Fast", "  FAST  "} {
+		if _, ok := Lookup(name); !ok {
+			t.Fatalf("Lookup(%q) should resolve", name)
+		}
+	}
+	if _, ok := Lookup("turbo"); ok {
+		t.Fatal("Lookup(turbo) should not resolve")
+	}
+}
+
+func TestNamesAreSorted(t *testing.T) {
+	want := []string{"balanced", "fast", "thorough"}
+	if got := Names(); !reflect.DeepEqual(got, want) {
+		t.Fatalf("Names() = %v, want %v", got, want)
+	}
+}
+
+// Escalation targets must be the displaced values only — the catalog carries
+// triggers, the call site supplies what was displaced, and nothing else may
+// appear in the escalation.
+func TestFastPolicyTargetsDisplacedValuesOnly(t *testing.T) {
+	policy := Fast.Policy(80, true)
+	if policy == nil {
+		t.Fatal("fast must arm an escalation policy")
+	}
+	if policy.Name != "fast" {
+		t.Fatalf("policy.Name = %q, want fast", policy.Name)
+	}
+	esc := policy.Escalate
+	if esc == nil {
+		t.Fatal("fast policy must carry an escalation")
+	}
+	if esc.MaxTurns != 80 {
+		t.Fatalf("Escalate.MaxTurns = %d, want the displaced 80", esc.MaxTurns)
+	}
+	if esc.ReasoningEffort != "" {
+		t.Fatalf("Escalate.ReasoningEffort = %q, want empty (displaced effort is the provider default by construction)", esc.ReasoningEffort)
+	}
+	if !esc.RestoreDefaultEffort {
+		t.Fatal("effort was profile-filled, so escalation must restore the provider default")
+	}
+	if esc.RestoreCompletionGate {
+		t.Fatal("fast does not touch the completion gate, so escalation must not either")
+	}
+	if esc.OnToolFailureStreak != Fast.EscalateOnToolFailureStreak ||
+		esc.OnCompletionUncertain != Fast.EscalateOnCompletionUncertain ||
+		esc.OnSelfCorrectFailure != Fast.EscalateOnSelfCorrectFailure ||
+		esc.OnRiskyMutation != Fast.EscalateOnRiskyMutation {
+		t.Fatalf("escalation triggers must mirror the catalog, got %+v", esc)
+	}
+}
+
+// The effort restore arms only when the profile actually filled the effort:
+// with an explicit user effort the profile backed off, so escalation must not
+// clear the user's choice.
+func TestFastPolicyExplicitEffortIsNeverRestored(t *testing.T) {
+	policy := Fast.Policy(80, false)
+	if policy == nil || policy.Escalate == nil {
+		t.Fatal("fast must still arm its triggers")
+	}
+	if policy.Escalate.RestoreDefaultEffort {
+		t.Fatal("the profile did not fill the effort, so escalation must leave it alone")
+	}
+}
+
+// A profile that did not displace the budget (explicit --max-turns pinned it)
+// must not let escalation move the ceiling at all.
+func TestFastPolicyZeroDisplacedLeavesCeilingUntouched(t *testing.T) {
+	policy := Fast.Policy(0, false)
+	if policy == nil || policy.Escalate == nil {
+		t.Fatal("fast must still arm its triggers with a zero displaced budget")
+	}
+	if policy.Escalate.MaxTurns != 0 {
+		t.Fatalf("Escalate.MaxTurns = %d, want 0 (no displaced value to restore)", policy.Escalate.MaxTurns)
+	}
+}
+
+// Thorough is already the maximum posture: no triggers, so no policy — the
+// loop must stay byte-identical to the same knobs set by hand.
+func TestThoroughPolicyIsNil(t *testing.T) {
+	if policy := Thorough.Policy(80, true); policy != nil {
+		t.Fatalf("thorough Policy must be nil, got %+v", policy)
+	}
+}
+
+// Pin the catalog's provisional values so a retune is a deliberate,
+// test-visible diff (these floors came from the Phase 0 baseline's read-class
+// evidence and are expected to move after the post-oracle-fix re-capture).
+func TestCatalogProvisionalValues(t *testing.T) {
+	if Fast.MaxTurns != 30 || Fast.ReasoningEffort != "low" || Fast.SelfCorrect {
+		t.Fatalf("fast knobs changed: %+v", Fast)
+	}
+	// Critical, not High: the sandbox classifies every shell command as at
+	// least high risk, so a High trigger would spend the one-shot escalation
+	// on the first benign `go test` of any coding task.
+	if Fast.EscalateOnRiskyMutation != sandbox.RiskCritical {
+		t.Fatalf("fast risky-mutation trigger = %q, want %q", Fast.EscalateOnRiskyMutation, sandbox.RiskCritical)
+	}
+	if Thorough.MaxTurns != 160 || Thorough.ReasoningEffort != "high" || !Thorough.SelfCorrect {
+		t.Fatalf("thorough knobs changed: %+v", Thorough)
+	}
+}

internal/mcp/oauth.go

@@ -133,6 +133,26 @@ func discoverAuthorizationServer(ctx context.Context, client *http.Client, baseU
 	}, nil
 }
 
+// validateResolvedEndpoints applies the shared https/loopback endpoint rule to
+// every non-empty endpoint in the resolved metadata — configured or discovered
+// alike — so discovery metadata can never downgrade an MCP login to an insecure
+// or attacker-controlled authorization, token, or registration endpoint.
+func validateResolvedEndpoints(metadata authServerMetadata) error {
+	for _, ep := range []struct{ kind, url string }{
+		{"authorization", metadata.AuthorizationEndpoint},
+		{"token", metadata.TokenEndpoint},
+		{"registration", metadata.RegistrationEndpoint},
+	} {
+		if strings.TrimSpace(ep.url) == "" {
+			continue
+		}
+		if err := oauth.ValidateEndpointURL(ep.url); err != nil {
+			return fmt.Errorf("mcp oauth: %s endpoint: %w", ep.kind, err)
+		}
+	}
+	return nil
+}
+
 // resolveAuthorizationServer discovers metadata and applies explicit config
 // overrides. Configured endpoints take precedence over discovered ones, and act
 // as a fallback when discovery fails or omits a value.
@@ -148,6 +168,9 @@ func resolveAuthorizationServer(ctx context.Context, client *http.Client, baseUR
 			TokenEndpoint:         strings.TrimSpace(cfg.TokenEndpoint),
 			RegistrationEndpoint:  strings.TrimSpace(cfg.RegistrationEndpoint),
 		}
+		if err := validateResolvedEndpoints(metadata); err != nil {
+			return authServerMetadata{}, err
+		}
 		return metadata, nil
 	}
 
@@ -179,6 +202,9 @@ func resolveAuthorizationServer(ctx context.Context, client *http.Client, baseUR
 	if strings.TrimSpace(metadata.TokenEndpoint) == "" {
 		return authServerMetadata{}, errors.New("no token endpoint discovered or configured")
 	}
+	if err := validateResolvedEndpoints(metadata); err != nil {
+		return authServerMetadata{}, err
+	}
 	return metadata, nil
 }
 

internal/mcp/oauth_test.go

@@ -73,6 +73,22 @@ func TestResolveEndpointsFallsBackToConfig(t *testing.T) {
 	}
 }
 
+// Discovered MCP metadata that serves an insecure endpoint must be rejected
+// before use, so discovery cannot downgrade the MCP login to an
+// attacker-controlled origin (issue #511).
+func TestResolveAuthorizationServerRejectsInsecureDiscovered(t *testing.T) {
+	mux := http.NewServeMux()
+	server := httptest.NewServer(mux)
+	defer server.Close()
+	mux.HandleFunc("/.well-known/oauth-authorization-server", func(w http.ResponseWriter, _ *http.Request) {
+		_, _ = w.Write([]byte(`{"issuer":"` + server.URL + `","token_endpoint":"` + server.URL + `/token","authorization_endpoint":"http://evil.example/authorize"}`))
+	})
+	_, err := resolveAuthorizationServer(context.Background(), server.Client(), server.URL, OAuthConfig{})
+	if err == nil || !strings.Contains(err.Error(), "authorization endpoint") {
+		t.Fatalf("err = %v, want insecure authorization endpoint rejection", err)
+	}
+}
+
 // roundTripperFunc adapts a function to http.RoundTripper so a test can
 // observe every outbound request.
 type roundTripperFunc func(*http.Request) (*http.Response, error)

internal/npmwrapper/npmwrapper_test.go

@@ -12,6 +12,54 @@ import (
 	"time"
 )
 
+// Mirror internal/cli/exec.go exit codes so wrapper doctor fallback tests assert
+// the same CLI contract as the Go doctor command.
+const (
+	wrapperExitSuccess = 0
+	wrapperExitUsage   = 2
+	wrapperExitDoctor  = 1
+)
+
+// Canonical Bun recovery copy from bin/zero.js bunRecoveryParagraph() — shared
+// by the generic missing-binary path and the doctor text fallback.
+const (
+	bunRecoveryLead        = "You installed with Bun, which does not run dependency lifecycle scripts"
+	bunPmTrustProject      = "bun pm trust @gitlawb/zero"
+	bunPmTrustGlobal       = "bun pm -g trust @gitlawb/zero"
+	bunRecoveryTrustedDeps = `"trustedDependencies": ["@gitlawb/zero"]`
+	buildFromSourceLead    = "If this platform has no prebuilt binary, build from source:"
+)
+
+func runWrapperFixture(t *testing.T, wrapperPath string, args ...string) (stdout string, stderr string, exitCode int) {
+	return runWrapperFixtureWithEnv(t, wrapperPath, nil, args...)
+}
+
+func runWrapperFixtureWithEnv(t *testing.T, wrapperPath string, extraEnv []string, args ...string) (stdout string, stderr string, exitCode int) {
+	t.Helper()
+	node := requireNode(t)
+	ctx, cancel := context.WithTimeout(context.Background(), nodeWrapperTimeout())
+	defer cancel()
+	command := nodeWrapperCommand(ctx, node, wrapperPath, args...)
+	command.Env = append(withoutEnvKey(command.Env, "ZERO_LOCAL_CONTROL_HELPERS"), "ZERO_LOCAL_CONTROL_HELPERS=")
+	command.Env = append(withoutEnvKey(command.Env, "ZERO_WRAPPER_SIMULATE_BUN"), extraEnv...)
+	var stdoutBuf, stderrBuf strings.Builder
+	command.Stdout = &stdoutBuf
+	command.Stderr = &stderrBuf
+	err := command.Run()
+	if ctx.Err() != nil {
+		t.Fatalf("wrapper timed out: %v; stdout: %s stderr: %s", ctx.Err(), stdoutBuf.String(), stderrBuf.String())
+	}
+	exitCode = wrapperExitSuccess
+	if err != nil {
+		exitErr, ok := err.(*exec.ExitError)
+		if !ok {
+			t.Fatalf("wrapper err = %v, want *exec.ExitError; stdout=%s stderr=%s", err, stdoutBuf.String(), stderrBuf.String())
+		}
+		exitCode = exitErr.ExitCode()
+	}
+	return stdoutBuf.String(), stderrBuf.String(), exitCode
+}
+
 func TestPackageBinPointsToNodeWrapper(t *testing.T) {
 	root := repoRoot(t)
 	bytes, err := os.ReadFile(filepath.Join(root, "package.json"))
@@ -393,6 +441,221 @@ func nodeArchName() string {
 	}
 }
 
+// Issue #405: `zero doctor` is the diagnostic command, so when the native
+// binary is the thing that's broken it must NOT bail with the generic wrapper
+// error; that's exactly the blind alley the bug report calls out. Instead it
+// emits a doctor-shaped FAIL line for the runtime so the user's own diagnostic
+// surfaces the real cause. We assert on the doctor-report shape (so the doctor
+// UX matches what the Go-side doctor.Format produces) and on exit 1 (a missing
+// binary is still a hard failure, not a pass).
+func TestNodeWrapperDoctorReportsMissingNativeBinaryAsDoctorFail(t *testing.T) {
+	wrapperPath := copyWrapperFixture(t)
+	stdout, stderr, exitCode := runWrapperFixture(t, wrapperPath, "doctor")
+	if exitCode != wrapperExitDoctor {
+		t.Fatalf("doctor exit = %d, want %d; stdout=%s stderr=%s", exitCode, wrapperExitDoctor, stdout, stderr)
+	}
+	if strings.TrimSpace(stderr) != "" {
+		t.Fatalf("doctor text report must go to stdout only, got stderr=%q", stderr)
+	}
+	// Doctor-shaped report, not the generic wrapper bail. Matches the shape the
+	// Go-side doctor.Format emits: a header, "Overall: <pass/fail>", then
+	// "[<status>] <id> - <message>" lines.
+	if !strings.Contains(stdout, "Zero doctor report (") {
+		t.Fatalf("doctor output should start with a doctor report header, got stdout=%q", stdout)
+	}
+	if !strings.Contains(stdout, "Overall: fail") {
+		t.Fatalf("doctor overall must be fail when the native binary is missing, got stdout=%q", stdout)
+	}
+	if !strings.Contains(stdout, "[fail] runtime.go") {
+		t.Fatalf("doctor must report a failing runtime.go check, got stdout=%q", stdout)
+	}
+	if !strings.Contains(stdout, "Native zero binary is missing next to the npm wrapper") {
+		t.Fatalf("doctor must name the actual cause (missing native binary), got stdout=%q", stdout)
+	}
+	// The actionable remedy must point at the postinstall script that would fix
+	// the install, not just "build from source".
+	if !strings.Contains(stdout, "postinstall.mjs") {
+		t.Fatalf("doctor remedy should name the postinstall script, got stdout=%q", stdout)
+	}
+	// Regression guard for the original blind-alley bug: the doctor path must
+	// NOT emit the generic wrapper bail (that's what sent users debugging the
+	// wrong thing).
+	if strings.Contains(stdout, "[zero] No native binary is available for this install") {
+		t.Fatalf("doctor must not emit the generic wrapper bail, got stdout=%q", stdout)
+	}
+}
+
+func TestNodeWrapperDoctorJSONReportsMissingNativeBinaryAsJSONFail(t *testing.T) {
+	wrapperPath := copyWrapperFixture(t)
+	stdout, stderr, exitCode := runWrapperFixture(t, wrapperPath, "doctor", "--json")
+	if exitCode != wrapperExitDoctor {
+		t.Fatalf("doctor --json exit = %d, want %d; stdout=%s stderr=%s", exitCode, wrapperExitDoctor, stdout, stderr)
+	}
+	if strings.TrimSpace(stderr) != "" {
+		t.Fatalf("doctor --json should write machine-readable output to stdout only, got stderr=%q", stderr)
+	}
+	var report struct {
+		GeneratedAt string `json:"generatedAt"`
+		OK          bool   `json:"ok"`
+		Checks      []struct {
+			ID      string         `json:"id"`
+			Label   string         `json:"label"`
+			Status  string         `json:"status"`
+			Message string         `json:"message"`
+			Details map[string]any `json:"details"`
+		} `json:"checks"`
+	}
+	if err := json.Unmarshal([]byte(stdout), &report); err != nil {
+		t.Fatalf("doctor --json stdout should be valid JSON, got %q: %v", stdout, err)
+	}
+	if report.GeneratedAt == "" {
+		t.Fatalf("doctor --json report should include generatedAt: %#v", report)
+	}
+	if report.OK {
+		t.Fatalf("doctor --json ok = true, want false: %#v", report)
+	}
+	if len(report.Checks) != 1 {
+		t.Fatalf("doctor --json checks length = %d, want 1: %#v", len(report.Checks), report.Checks)
+	}
+	check := report.Checks[0]
+	if check.ID != "runtime.go" || check.Label != "Go runtime" || check.Status != "fail" {
+		t.Fatalf("doctor --json runtime check = %#v, want failing runtime.go check", check)
+	}
+	if !strings.Contains(check.Message, "Native zero binary is missing next to the npm wrapper") {
+		t.Fatalf("doctor --json must name the actual cause, got %#v", check)
+	}
+	remedy, _ := check.Details["remedy"].(string)
+	if !strings.Contains(remedy, "postinstall.mjs") {
+		t.Fatalf("doctor --json remedy should name the postinstall script, got %#v", check.Details)
+	}
+}
+
+func TestNodeWrapperDoctorHelpShowsUsage(t *testing.T) {
+	wrapperPath := copyWrapperFixture(t)
+	for _, args := range [][]string{{"doctor", "--help"}, {"doctor", "help"}, {"doctor", "-h"}} {
+		stdout, stderr, exitCode := runWrapperFixture(t, wrapperPath, args...)
+		if exitCode != wrapperExitSuccess {
+			t.Fatalf("%v exit = %d, want %d; stdout=%s stderr=%s", args, exitCode, wrapperExitSuccess, stdout, stderr)
+		}
+		if strings.TrimSpace(stderr) != "" {
+			t.Fatalf("%v help must write to stdout only, got stderr=%q", args, stderr)
+		}
+		if !strings.Contains(stdout, "Usage:") || !strings.Contains(stdout, "zero doctor [flags]") {
+			t.Fatalf("%v help output = %q, want doctor usage text", args, stdout)
+		}
+	}
+}
+
+func TestNodeWrapperDoctorRejectsUnknownFlag(t *testing.T) {
+	wrapperPath := copyWrapperFixture(t)
+	stdout, stderr, exitCode := runWrapperFixture(t, wrapperPath, "doctor", "--bogus")
+	if exitCode != wrapperExitUsage {
+		t.Fatalf("doctor --bogus exit = %d, want %d; stdout=%s stderr=%s", exitCode, wrapperExitUsage, stdout, stderr)
+	}
+	if strings.TrimSpace(stdout) != "" {
+		t.Fatalf("invalid doctor flag should not write stdout, got %q", stdout)
+	}
+	if stderr != "[zero] unknown doctor flag \"--bogus\"\n" {
+		t.Fatalf("invalid doctor flag stderr = %q", stderr)
+	}
+}
+
+// `doctor --connectivity` (a valid doctor invocation with a trailing flag) must
+// take the same doctor-shaped path: parseDoctorArgs accepts --connectivity and
+// the missing-binary fallback still reports the runtime failure.
+func TestNodeWrapperDoctorWithFlagsStillReportsDoctorFail(t *testing.T) {
+	wrapperPath := copyWrapperFixture(t)
+	stdout, stderr, exitCode := runWrapperFixture(t, wrapperPath, "doctor", "--connectivity")
+	if exitCode != wrapperExitDoctor {
+		t.Fatalf("doctor --connectivity exit = %d, want %d; stdout=%s stderr=%s", exitCode, wrapperExitDoctor, stdout, stderr)
+	}
+	if strings.TrimSpace(stderr) != "" {
+		t.Fatalf("doctor --connectivity must write to stdout only, got stderr=%q", stderr)
+	}
+	if !strings.Contains(stdout, "[fail] runtime.go") {
+		t.Fatalf("doctor --connectivity must still emit the failing runtime.go line, got stdout=%q", stdout)
+	}
+	if strings.Contains(stdout, "[zero] No native binary is available for this install") {
+		t.Fatalf("doctor --connectivity must not fall back to the generic bail, got stdout=%q", stdout)
+	}
+}
+
+func TestNodeWrapperGenericReportsBunRecoveryWhenInstalledByBun(t *testing.T) {
+	wrapperPath := copyWrapperFixture(t)
+	_, stderr, exitCode := runWrapperFixtureWithEnv(t, wrapperPath, []string{"ZERO_WRAPPER_SIMULATE_BUN=1"}, "--version")
+	if exitCode != wrapperExitDoctor {
+		t.Fatalf("generic missing-binary exit = %d, want %d; stderr=%s", exitCode, wrapperExitDoctor, stderr)
+	}
+	assertBunRecoveryMessage(t, stderr)
+	if !strings.Contains(stderr, buildFromSourceLead) {
+		t.Fatalf("generic bun stderr should include build-from-source guidance, got %q", stderr)
+	}
+}
+
+func TestNodeWrapperDoctorReportsBunRecoveryWhenInstalledByBun(t *testing.T) {
+	wrapperPath := copyWrapperFixture(t)
+	stdout, stderr, exitCode := runWrapperFixtureWithEnv(t, wrapperPath, []string{"ZERO_WRAPPER_SIMULATE_BUN=1"}, "doctor")
+	if exitCode != wrapperExitDoctor {
+		t.Fatalf("doctor bun exit = %d, want %d; stdout=%s stderr=%s", exitCode, wrapperExitDoctor, stdout, stderr)
+	}
+	if strings.TrimSpace(stderr) != "" {
+		t.Fatalf("doctor bun report must go to stdout only, got stderr=%q", stderr)
+	}
+	if !strings.Contains(stdout, "[fail] runtime.go") {
+		t.Fatalf("doctor bun report must still fail runtime.go, got stdout=%q", stdout)
+	}
+	assertBunRecoveryMessage(t, stdout)
+	if !strings.Contains(stdout, buildFromSourceLead) {
+		t.Fatalf("doctor bun report should reuse generic build-from-source guidance, got stdout=%q", stdout)
+	}
+	if strings.Contains(stdout, "If reinstall fails") {
+		t.Fatalf("doctor bun report must not use a separate build-from-source string, got stdout=%q", stdout)
+	}
+}
+
+func TestNodeWrapperDoctorAndGenericShareBunRecoveryCopy(t *testing.T) {
+	wrapperPath := copyWrapperFixture(t)
+	simulateBun := []string{"ZERO_WRAPPER_SIMULATE_BUN=1"}
+	_, genericStderr, _ := runWrapperFixtureWithEnv(t, wrapperPath, simulateBun, "--version")
+	doctorStdout, _, _ := runWrapperFixtureWithEnv(t, wrapperPath, simulateBun, "doctor")
+	genericBun := extractBunRecoveryBlock(genericStderr)
+	doctorBun := extractBunRecoveryBlock(doctorStdout)
+	if genericBun != doctorBun {
+		t.Fatalf("bun recovery copy drift:\ngeneric=%q\ndoctor=%q", genericBun, doctorBun)
+	}
+}
+
+func assertBunRecoveryMessage(t *testing.T, output string) {
+	t.Helper()
+	if !strings.Contains(output, bunRecoveryLead) {
+		t.Fatalf("output should include bun recovery lead, got %q", output)
+	}
+	if !strings.Contains(output, bunPmTrustProject) {
+		t.Fatalf("output should include project bun pm trust guidance, got %q", output)
+	}
+	if !strings.Contains(output, bunPmTrustGlobal) {
+		t.Fatalf("output should include global bun pm trust guidance, got %q", output)
+	}
+	if !strings.Contains(output, bunRecoveryTrustedDeps) {
+		t.Fatalf("output should include trustedDependencies fallback guidance, got %q", output)
+	}
+	if !strings.Contains(output, "to your project package.json and reinstall.") {
+		t.Fatalf("output should mention package.json reinstall fallback, got %q", output)
+	}
+}
+
+func extractBunRecoveryBlock(output string) string {
+	start := strings.Index(output, bunRecoveryLead)
+	if start < 0 {
+		return ""
+	}
+	end := strings.Index(output[start:], buildFromSourceLead)
+	if end < 0 {
+		return strings.TrimSpace(output[start:])
+	}
+	return strings.TrimSpace(output[start : start+end])
+}
+
 func TestNodeWrapperLaunchesNativeBinary(t *testing.T) {
 	if runtime.GOOS == "windows" {
 		t.Skip("mock executable fixture uses a POSIX shell script")

internal/oauth/device.go

@@ -49,9 +49,9 @@ func RequestDeviceCode(ctx context.Context, client *http.Client, cfg Config, now
 	if err := validateTokenEndpoint(endpoint); err != nil {
 		return DeviceAuth{}, err
 	}
-	if client == nil {
-		client = http.DefaultClient
-	}
+	// Refuse redirects so a device-authorization redirect can't replay the
+	// client_id/client_secret POST body to an unvalidated origin.
+	client = withoutRedirects(client)
 	if now == nil {
 		now = time.Now
 	}
@@ -157,9 +157,9 @@ func pollDeviceOnce(ctx context.Context, client *http.Client, cfg Config, device
 	if err := validateTokenEndpoint(cfg.TokenEndpoint); err != nil {
 		return Token{}, err
 	}
-	if client == nil {
-		client = http.DefaultClient
-	}
+	// Refuse redirects so a poll redirect can't replay the device_code/secret
+	// POST body to an unvalidated origin.
+	client = withoutRedirects(client)
 	form := url.Values{}
 	form.Set("grant_type", deviceGrantType)
 	form.Set("device_code", deviceCode)

internal/oauth/device_test.go

@@ -2,6 +2,7 @@ package oauth
 
 import (
 	"context"
+	"errors"
 	"net/http"
 	"net/http/httptest"
 	"strings"
@@ -28,6 +29,30 @@ func TestRequestDeviceCodeSendsClientSecret(t *testing.T) {
 	}
 }
 
+func TestRequestDeviceCodeRefusesRedirect(t *testing.T) {
+	endpoint, hits := redirectTrap(t, http.StatusTemporaryRedirect) // 307
+	_, err := RequestDeviceCode(context.Background(), http.DefaultClient,
+		Config{ClientID: "c", ClientSecret: "shh", DeviceAuthorizationEndpoint: endpoint}, nil)
+	if !errors.Is(err, ErrUnsafeRedirect) {
+		t.Fatalf("RequestDeviceCode err = %v, want ErrUnsafeRedirect", err)
+	}
+	if n := hits(); n != 0 {
+		t.Fatalf("attacker received %d request(s) — client_secret replayed", n)
+	}
+}
+
+func TestPollDeviceOnceRefusesRedirect(t *testing.T) {
+	endpoint, hits := redirectTrap(t, http.StatusPermanentRedirect) // 308
+	_, err := pollDeviceOnce(context.Background(), http.DefaultClient,
+		Config{ClientID: "c", ClientSecret: "shh", TokenEndpoint: endpoint}, "device-code", nil)
+	if !errors.Is(err, ErrUnsafeRedirect) {
+		t.Fatalf("pollDeviceOnce err = %v, want ErrUnsafeRedirect", err)
+	}
+	if n := hits(); n != 0 {
+		t.Fatalf("attacker received %d request(s) — device_code/secret replayed", n)
+	}
+}
+
 func TestRequestDeviceCode(t *testing.T) {
 	server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, _ *http.Request) {
 		_, _ = w.Write([]byte(`{"device_code":"dc","user_code":"WXYZ-1234","verification_uri":"https://x/dev","verification_uri_complete":"https://x/dev?c=WXYZ","expires_in":600,"interval":3}`))

internal/oauth/flow.go

@@ -39,6 +39,13 @@ func BuildAuthorizationURL(cfg Config, pkce PKCE, state, redirectURI string, ext
 	if endpoint == "" {
 		return "", errors.New("oauth: no authorization endpoint configured")
 	}
+	// Backstop: validate the authorization endpoint at the shared choke point
+	// both the provider and MCP flows build their browser URL through, so a
+	// discovery-downgraded endpoint can never open in the browser even if a
+	// merge site missed it.
+	if err := ValidateEndpointURL(endpoint); err != nil {
+		return "", err
+	}
 	parsed, err := url.Parse(endpoint)
 	if err != nil {
 		return "", fmt.Errorf("oauth: parse authorization endpoint: %w", err)
@@ -83,13 +90,15 @@ func isReservedAuthParam(key string) bool {
 	}
 }
 
-// validateTokenEndpoint refuses to send a credential to a token endpoint that is
-// not https, unless it is a loopback host (mirrors oauth2-client.js
-// validateTokenEndpointURL). This prevents leaking codes/tokens over cleartext.
-func validateTokenEndpoint(endpoint string) error {
+// ValidateEndpointURL refuses an OAuth endpoint that is not https, unless it is
+// a loopback host (mirrors oauth2-client.js validateTokenEndpointURL). Every
+// credential-bearing endpoint — configured OR learned from discovery — must
+// pass this single rule, so discovery metadata can never downgrade a login to
+// cleartext or redirect it to an attacker-controlled http origin.
+func ValidateEndpointURL(endpoint string) error {
 	parsed, err := url.Parse(trimmed(endpoint))
 	if err != nil || parsed.Host == "" {
-		return fmt.Errorf("oauth: invalid token endpoint %q", endpoint)
+		return fmt.Errorf("oauth: invalid endpoint %q", endpoint)
 	}
 	if parsed.Scheme == "https" {
 		return nil
@@ -100,6 +109,30 @@ func validateTokenEndpoint(endpoint string) error {
 	return fmt.Errorf("%w: %s", ErrInsecureTokenEndpoint, parsed.Scheme+"://"+parsed.Host)
 }
 
+// validateTokenEndpoint checks a token endpoint against the shared endpoint rule
+// (kept as a named helper for the token-exchange and device call sites). This
+// prevents leaking codes/tokens over cleartext.
+func validateTokenEndpoint(endpoint string) error {
+	return ValidateEndpointURL(endpoint)
+}
+
+// withoutRedirects returns a shallow copy of client whose redirect policy
+// refuses to follow ANY redirect. A credential-bearing OAuth POST (token
+// exchange/refresh, device authorization, device-token poll) must never let a
+// 307/308 replay its form body — codes, PKCE verifiers, refresh tokens, client
+// secrets — to a target that was never endpoint-validated. The caller's client
+// is left unmodified.
+func withoutRedirects(client *http.Client) *http.Client {
+	if client == nil {
+		client = http.DefaultClient
+	}
+	copied := *client
+	copied.CheckRedirect = func(*http.Request, []*http.Request) error {
+		return ErrUnsafeRedirect
+	}
+	return &copied
+}
+
 func isLoopbackHost(host string) bool {
 	if host == "localhost" {
 		return true
@@ -160,9 +193,9 @@ func PostToken(ctx context.Context, client *http.Client, tokenEndpoint string, f
 	if err := validateTokenEndpoint(tokenEndpoint); err != nil {
 		return Token{}, err
 	}
-	if client == nil {
-		client = http.DefaultClient
-	}
+	// Refuse redirects so a 307/308 from the token endpoint can't replay this
+	// credential-bearing POST body to another (unvalidated) origin.
+	client = withoutRedirects(client)
 	if now == nil {
 		now = time.Now
 	}

internal/oauth/flow_test.go

@@ -5,10 +5,12 @@ import (
 	"crypto/sha256"
 	"encoding/base64"
 	"errors"
+	"io"
 	"net/http"
 	"net/http/httptest"
 	"net/url"
 	"strings"
+	"sync/atomic"
 	"testing"
 	"time"
 )
@@ -112,6 +114,78 @@ func TestBuildAuthorizationURLRejectsPlain(t *testing.T) {
 	}
 }
 
+// The shared choke point both flows build their browser URL through must refuse
+// an insecure authorization endpoint, so a discovery-downgraded endpoint can
+// never open in the browser even if a merge site missed it (issue #511).
+func TestBuildAuthorizationURLRejectsInsecureEndpoint(t *testing.T) {
+	_, err := BuildAuthorizationURL(
+		Config{AuthorizationEndpoint: "http://evil.example/authorize", ClientID: "c"},
+		PKCE{Method: MethodS256, Challenge: "chal"},
+		"state", "http://127.0.0.1/cb", nil,
+	)
+	if !errors.Is(err, ErrInsecureTokenEndpoint) {
+		t.Fatalf("err = %v, want ErrInsecureTokenEndpoint", err)
+	}
+}
+
+// redirectTrap starts an "attacker" server that counts any hit, and a credential
+// endpoint that redirects to it with the given status. It returns the endpoint
+// URL and a func reporting how many times the attacker was contacted — a
+// non-zero count means the credential-bearing POST body was replayed (#729).
+func redirectTrap(t *testing.T, status int) (endpoint string, attackerHits func() int32) {
+	t.Helper()
+	var hits int32
+	attacker := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, _ *http.Request) {
+		atomic.AddInt32(&hits, 1)
+		_, _ = io.WriteString(w, `{"access_token":"leaked","refresh_token":"leaked"}`)
+	}))
+	t.Cleanup(attacker.Close)
+	endp := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		http.Redirect(w, r, attacker.URL, status)
+	}))
+	t.Cleanup(endp.Close)
+	return endp.URL, func() int32 { return atomic.LoadInt32(&hits) }
+}
+
+func TestPostTokenRefusesRedirect(t *testing.T) {
+	endpoint, hits := redirectTrap(t, http.StatusTemporaryRedirect) // 307
+	form := url.Values{}
+	form.Set("grant_type", "authorization_code")
+	form.Set("code", "the-code")
+	form.Set("client_secret", "shh")
+	_, err := PostToken(context.Background(), http.DefaultClient, endpoint, form, Token{}, nil)
+	if !errors.Is(err, ErrUnsafeRedirect) {
+		t.Fatalf("PostToken err = %v, want ErrUnsafeRedirect", err)
+	}
+	if n := hits(); n != 0 {
+		t.Fatalf("attacker received %d request(s) — credential body was replayed", n)
+	}
+}
+
+func TestExchangeCodeRefusesRedirect(t *testing.T) {
+	endpoint, hits := redirectTrap(t, http.StatusPermanentRedirect) // 308
+	_, err := ExchangeCode(context.Background(), http.DefaultClient,
+		Config{ClientID: "c", TokenEndpoint: endpoint}, "the-code", "verifier", "http://127.0.0.1/cb", nil)
+	if !errors.Is(err, ErrUnsafeRedirect) {
+		t.Fatalf("ExchangeCode err = %v, want ErrUnsafeRedirect", err)
+	}
+	if n := hits(); n != 0 {
+		t.Fatalf("attacker received %d request(s) — code/verifier replayed", n)
+	}
+}
+
+func TestRefreshRefusesRedirect(t *testing.T) {
+	endpoint, hits := redirectTrap(t, http.StatusTemporaryRedirect) // 307
+	_, err := Refresh(context.Background(), http.DefaultClient,
+		Config{ClientID: "c", TokenEndpoint: endpoint}, Token{RefreshToken: "rt"}, nil)
+	if !errors.Is(err, ErrUnsafeRedirect) {
+		t.Fatalf("Refresh err = %v, want ErrUnsafeRedirect", err)
+	}
+	if n := hits(); n != 0 {
+		t.Fatalf("attacker received %d request(s) — refresh_token replayed", n)
+	}
+}
+
 func TestValidateTokenEndpoint(t *testing.T) {
 	cases := map[string]bool{
 		"https://auth.example.com/token": true,

internal/oauth/manager.go

@@ -218,13 +218,26 @@ func (m *Manager) resolveEndpoints(ctx context.Context, cfg Config) (Config, err
 	// may already be sufficient for the chosen flow. Only merge on success (and
 	// never return from the error branch, which would trip the nilerr linter).
 	if meta, err := m.discover(ctx, cfg.IssuerURL); err == nil {
-		if cfg.AuthorizationEndpoint == "" {
+		// A discovered endpoint must pass the same https/loopback rule as a
+		// configured one before it is merged, so discovery metadata can never
+		// silently downgrade the login to an insecure or attacker-controlled
+		// endpoint. Fail closed on a bad endpoint rather than merging it.
+		if cfg.AuthorizationEndpoint == "" && meta.AuthorizationEndpoint != "" {
+			if err := ValidateEndpointURL(meta.AuthorizationEndpoint); err != nil {
+				return cfg, fmt.Errorf("oauth: discovered authorization endpoint: %w", err)
+			}
 			cfg.AuthorizationEndpoint = meta.AuthorizationEndpoint
 		}
-		if cfg.TokenEndpoint == "" {
+		if cfg.TokenEndpoint == "" && meta.TokenEndpoint != "" {
+			if err := ValidateEndpointURL(meta.TokenEndpoint); err != nil {
+				return cfg, fmt.Errorf("oauth: discovered token endpoint: %w", err)
+			}
 			cfg.TokenEndpoint = meta.TokenEndpoint
 		}
-		if cfg.DeviceAuthorizationEndpoint == "" {
+		if cfg.DeviceAuthorizationEndpoint == "" && meta.DeviceAuthorizationEndpoint != "" {
+			if err := ValidateEndpointURL(meta.DeviceAuthorizationEndpoint); err != nil {
+				return cfg, fmt.Errorf("oauth: discovered device authorization endpoint: %w", err)
+			}
 			cfg.DeviceAuthorizationEndpoint = meta.DeviceAuthorizationEndpoint
 		}
 	}

internal/oauth/manager_test.go

@@ -2,6 +2,7 @@ package oauth
 
 import (
 	"context"
+	"errors"
 	"io"
 	"net/http"
 	"net/http/httptest"
@@ -200,6 +201,24 @@ func TestManagerGetFreshDiscoversIssuerEndpoints(t *testing.T) {
 	}
 }
 
+// A discovery document that serves an insecure (non-https, non-loopback)
+// endpoint must be rejected before it is merged, so discovery metadata cannot
+// silently downgrade the login to an attacker-controlled origin (issue #511).
+func TestResolveEndpointsRejectsInsecureDiscoveredEndpoint(t *testing.T) {
+	mux := http.NewServeMux()
+	server := httptest.NewServer(mux)
+	defer server.Close()
+	mux.HandleFunc("/.well-known/oauth-authorization-server", func(w http.ResponseWriter, _ *http.Request) {
+		// Loopback issuer, but an attacker-controlled http authorization endpoint.
+		_, _ = io.WriteString(w, `{"issuer":"`+server.URL+`","token_endpoint":"`+server.URL+`/token","authorization_endpoint":"http://evil.example/authorize"}`)
+	})
+	m := managerFor(t, map[string]string{}, nil)
+	_, err := m.resolveEndpoints(context.Background(), Config{IssuerURL: server.URL})
+	if !errors.Is(err, ErrInsecureTokenEndpoint) {
+		t.Fatalf("resolveEndpoints err = %v, want ErrInsecureTokenEndpoint", err)
+	}
+}
+
 func TestManagerGetFreshRefreshesExpired(t *testing.T) {
 	fp := newFakeProvider(t, `{"access_token":"fresh-at","expires_in":3600}`)
 	env := map[string]string{

internal/oauth/oauth.go

@@ -90,6 +90,11 @@ var (
 	// ErrInsecureTokenEndpoint is returned when a credential would be sent over a
 	// non-https, non-loopback endpoint.
 	ErrInsecureTokenEndpoint = errors.New("oauth: refusing to send credential to a non-https token endpoint")
+	// ErrUnsafeRedirect is returned when a credential-bearing OAuth POST (token
+	// exchange/refresh, device authorization, or device-token poll) is redirected.
+	// Following it would replay the form body — codes, PKCE verifiers, refresh
+	// tokens, client secrets — to a target that never passed endpoint validation.
+	ErrUnsafeRedirect = errors.New("oauth: refusing to follow a redirect from a credential endpoint")
 	// ErrNoRefreshToken is returned when a refresh is attempted without one.
 	ErrNoRefreshToken = errors.New("oauth: no refresh token available")
 	// ErrAuthorizationPending is the RFC 8628 "keep polling" signal.

internal/perfbench/manifests/baseline.json

@@ -133,7 +133,7 @@
       "verificationCommand": [
         "bash",
         "-c",
-        "go test ./... && ! grep -RIn 'MaxRetries =' ."
+        "go test ./... && ! grep --exclude=.zero-answer.txt -RIn 'MaxRetries =' ."
       ],
       "oracleTest": "package edit\n\nvar _ = RetryLimit\n"
     },
@@ -146,7 +146,7 @@
       "verificationCommand": [
         "bash",
         "-c",
-        "go test ./... && ! grep -R receieves . && grep -R receives ."
+        "go test ./... && ! grep --exclude=.zero-answer.txt -R receieves . && grep --exclude=.zero-answer.txt -R receives ."
       ]
     },
     {
@@ -400,7 +400,7 @@
       "verificationCommand": [
         "bash",
         "-c",
-        "go test ./... && ! grep -R 'package refactor' . && grep -R 'package zeroapp' ."
+        "go test ./... && ! grep --exclude=.zero-answer.txt -R 'package refactor' . && grep --exclude=.zero-answer.txt -R 'package zeroapp' ."
       ],
       "oracleTest": "package zeroapp\n\nvar _ = Config{}\n"
     },
@@ -426,7 +426,7 @@
       "verificationCommand": [
         "bash",
         "-c",
-        "go test ./... && ! grep -R 'func Wrapper' ."
+        "go test ./... && ! grep --exclude=.zero-answer.txt -R 'func Wrapper' ."
       ],
       "oracleTest": "package refactor\n\nimport \"testing\"\n\n// TestGreetWrapped proves greetWrapped still exists and behaves as the inlined\n// Wrapper did: deleting both Wrapper and greetWrapped (or stubbing greetWrapped\n// to return \"\") fails to compile or run, so the oracle can't be passed by\n// removing the target along with its caller. Paired with the ! grep 'func\n// Wrapper' verificationCommand that proves Wrapper itself is gone.\nfunc TestGreetWrapped(t *testing.T) {\n\tif got := greetWrapped(\"x\"); got != \"hello, x\" {\n\t\tt.Fatalf(\"greetWrapped(x) = %q, want %q\", got, \"hello, x\")\n\t}\n}\n"
     },

internal/perfbench/taskbench.go

@@ -85,6 +85,10 @@ type RunContext struct {
 	Model       string
 	Mode        string
 	SelfCorrect bool
+	// ExecProfile, when non-empty, is forwarded to the agent invocation as
+	// --exec-profile (turn benchmark runner; the task benchmark leaves it
+	// empty).
+	ExecProfile string
 }
 
 // TaskRunner runs a single benchmark task and returns its outcome.

internal/perfbench/turn_bench.go

@@ -15,6 +15,7 @@ import (
 	"strings"
 	"time"
 
+	"github.com/Gitlawb/zero/internal/execprofile"
 	"github.com/Gitlawb/zero/internal/trace"
 )
 
@@ -40,7 +41,25 @@ import (
 // prevents a v2→v2 cross-version comparison from misreading the jump as a model
 // improvement. New counts: correctness 34 (edit 10 + fix 8 + nav 10 + refactor
 // 6), build 0, latency 14 (longproc 4 + longctx 4 + parallel 6).
-const TurnSchemaVersion = 3
+//
+// v4 adds tasksErrored: the number of tasks whose every iteration produced no
+// accepted benchmark sample. That covers pre-run failures (missing binary,
+// spawn error, process crash) and harness errors after a successful agent run
+// (e.g. oracle stamping failed) alike — in every case the iteration yields no
+// usable measurement. Errored tasks were previously visible only in warnings,
+// so a run where nothing was actually measured still printed a clean-looking
+// 0% pass summary and exited 0. tasksErrored makes that state first-class:
+// the summary surfaces it, and the turn command exits nonzero when every
+// attempted task errored. Accounting: an errored oracle task stays in its
+// tier denominator as a failure; an errored latency-only task counts in
+// latencyOnlyTasks and, as always, in no pass rate.
+//
+// v5 is ADDITIVE ONLY: execProfile records the execution profile the run was
+// benchmarked under (omitted when none was selected), so profile A/B captures
+// are self-describing and two reports can't be compared without noticing they
+// ran different postures. No existing field changed shape or meaning; a v4
+// consumer reading a v5 report misses only the new field.
+const TurnSchemaVersion = 5
 
 // TurnRunner runs one benchmark task and reports its outcome plus the captured
 // per-turn trace. A non-nil Err means the run failed to execute (process crash);
@@ -67,6 +86,10 @@ type TurnBenchConfig struct {
 	Model       string
 	Mode        string
 	SelfCorrect bool
+	// ExecProfile, when non-empty, runs every task under the named execution
+	// profile (zero exec --exec-profile) and stamps it into the result so the
+	// report is self-describing for profile A/B comparisons.
+	ExecProfile string
 	Version     string
 	Commit      string
 	// Iterations is how many times each task is run. The per-process `zero exec`
@@ -139,11 +162,13 @@ type TurnBenchResult struct {
 	Suite               string                  `json:"suite"`
 	Model               string                  `json:"model"`
 	Mode                string                  `json:"mode,omitempty"`
+	ExecProfile         string                  `json:"execProfile,omitempty"`
 	SelfCorrect         bool                    `json:"selfCorrect"`
 	Version             string                  `json:"version,omitempty"`
 	Commit              string                  `json:"commit,omitempty"`
 	Date                string                  `json:"date"`
 	TasksAttempted      int                     `json:"tasksAttempted"`
+	TasksErrored        int                     `json:"tasksErrored"`
 	TasksVerified       int                     `json:"tasksVerified"`
 	TasksPassed         int                     `json:"tasksPassed"`
 	LatencyOnlyTasks    int                     `json:"latencyOnlyTasks"`
@@ -189,6 +214,19 @@ func RunTurnBench(ctx context.Context, set TaskSet, cfg TurnBenchConfig) (TurnBe
 	if cfg.Runner == nil {
 		return TurnBenchResult{}, errors.New("turn benchmark requires a runner")
 	}
+	// Canonicalize the profile once at the boundary so every consumer — the
+	// child exec args, the stamped result, the summary — carries the same
+	// catalog name regardless of the caller's casing/whitespace, and a direct
+	// library caller cannot slip an unknown name into a report the way the CLI
+	// (which validates at parse time) cannot.
+	benchProfile := strings.TrimSpace(cfg.ExecProfile)
+	if benchProfile != "" {
+		profile, ok := execprofile.Lookup(benchProfile)
+		if !ok {
+			return TurnBenchResult{}, fmt.Errorf("unknown execution profile %q (valid: %s)", cfg.ExecProfile, strings.Join(execprofile.Names(), ", "))
+		}
+		benchProfile = profile.Name
+	}
 	iterations := cfg.Iterations
 	if iterations < 1 {
 		iterations = 1
@@ -197,7 +235,7 @@ func RunTurnBench(ctx context.Context, set TaskSet, cfg TurnBenchConfig) (TurnBe
 	if now == nil {
 		now = time.Now
 	}
-	rc := RunContext{Model: cfg.Model, Mode: cfg.Mode, SelfCorrect: cfg.SelfCorrect}
+	rc := RunContext{Model: cfg.Model, Mode: cfg.Mode, SelfCorrect: cfg.SelfCorrect, ExecProfile: benchProfile}
 
 	perSpanSamples := map[string][]float64{}
 	classWalls := map[string][]float64{}
@@ -227,6 +265,7 @@ func RunTurnBench(ctx context.Context, set TaskSet, cfg TurnBenchConfig) (TurnBe
 		Suite:         strings.TrimSpace(set.ID),
 		Model:         strings.TrimSpace(cfg.Model),
 		Mode:          strings.TrimSpace(cfg.Mode),
+		ExecProfile:   benchProfile,
 		SelfCorrect:   cfg.SelfCorrect,
 		Version:       strings.TrimSpace(cfg.Version),
 		Commit:        strings.TrimSpace(cfg.Commit),
@@ -249,6 +288,7 @@ func RunTurnBench(ctx context.Context, set TaskSet, cfg TurnBenchConfig) (TurnBe
 		// runner is cold-start, so a flaky pass on one iteration and a fail on
 		// another is a real regression signal, not noise to average away.
 		passedForTask := true
+		erroredIterations := 0
 		for iter := 0; iter < iterations; iter++ {
 			outcome := cfg.Runner(ctx, task, rc)
 			if outcome.TraceIssue != "" {
@@ -266,6 +306,7 @@ func RunTurnBench(ctx context.Context, set TaskSet, cfg TurnBenchConfig) (TurnBe
 					Message: fmt.Sprintf("task %s: %v", task.ID, outcome.Err),
 				})
 				passedForTask = false
+				erroredIterations++
 				continue
 			}
 			if !outcome.Passed {
@@ -306,6 +347,14 @@ func RunTurnBench(ctx context.Context, set TaskSet, cfg TurnBenchConfig) (TurnBe
 		// counters. A latency-only task (no verificationCommand) is never counted
 		// in any pass rate even when the runner reports Passed — an exit-0
 		// read-only run proves the turn ran, not that the answer was right.
+		if erroredIterations == iterations {
+			// Every iteration produced no accepted sample (spawn failure or a
+			// harness error after the run). The task still counts in its tier
+			// below — an oracle task as a failure, a latency-only task in no
+			// pass rate as always — but the errored state is first-class so a
+			// broken run can never print a clean-looking summary.
+			result.TasksErrored++
+		}
 		hasOracle := len(task.VerificationCommand) > 0
 		switch {
 		case !hasOracle:
@@ -456,9 +505,26 @@ func FormatTurnBenchSummary(result TurnBenchResult) string {
 			result.BuildPassedTasks, result.BuildCheckedTasks, result.BuildPassRate*100,
 			result.LatencyOnlyTasks, result.Iterations),
 	}
+	if result.TasksErrored > 0 {
+		lines = append(lines, fmt.Sprintf("ERRORED: %d task(s) produced no accepted benchmark sample (spawn/crash or harness error) — errored oracle tasks count as failures in their tier's pass rate; latency-only tasks stay out of pass rates as always", result.TasksErrored))
+		shown := 0
+		for _, warning := range result.Warnings {
+			if warning.Metric != "run" {
+				continue
+			}
+			lines = append(lines, "  "+warning.Message)
+			shown++
+			if shown == 3 {
+				break
+			}
+		}
+	}
 	if result.Mode != "" {
 		lines = append(lines, "mode: "+result.Mode)
 	}
+	if result.ExecProfile != "" {
+		lines = append(lines, "exec-profile: "+result.ExecProfile)
+	}
 	if len(result.TopLatency) > 0 {
 		lines = append(lines, "top latency sources:")
 		for _, src := range result.TopLatency {
@@ -512,28 +578,54 @@ func WriteTurnBenchJSON(w io.Writer, result TurnBenchResult) error {
 	return encoder.Encode(result)
 }
 
+// execExitIncomplete mirrors internal/cli.exitIncomplete (the `zero exec` exit
+// code, 4) for a headless run the completion gate marked INCOMPLETE: the run
+// stopped without a completion signal (e.g. it couldn't self-verify because its
+// sandboxed shell was unavailable under --auto member), which is distinct from a
+// crash (1), usage error (2), provider failure (3), or interruption (130). It is
+// the ONLY nonzero exit an oracle-bearing task may defer to the oracle for; every
+// other nonzero exit stays authoritative. Kept as a local literal so perfbench
+// doesn't import the cli package; must stay in sync with that constant.
+const execExitIncomplete = 4
+
 // NewTurnExecRunner builds the production turn-benchmark runner: it invokes
 // headless `zero exec` with stream-json output AND `--trace <tmpfile>`, then
 // parses the emitted NDJSON trace into a *trace.TurnTrace. binary is the path to
 // the `zero` binary; extraArgs are appended to every invocation. Pass/fail is
-// decided from the stream-json run_end exit code (and the task's
-// VerificationCommand when present), exactly like NewExecRunner.
+// decided by the task's oracle when it has one. The oracle that gates the verdict
+// is the VerificationCommand (which compiles and runs the stamped OracleTest via
+// `go test`, plus any greps against the fixture); it is ground truth, so an
+// oracle-bearing task that applied the right edit but exited INCOMPLETE (exit 4,
+// its sandboxed self-verification couldn't run) still passes. Any OTHER nonzero
+// exit (crash/usage/provider/interrupt) stays authoritative and fails even an
+// oracle-bearing task. A latency-only task has no VerificationCommand, so any
+// nonzero exit is its only signal and fails it.
 func NewTurnExecRunner(binary string, extraArgs ...string) TurnRunner {
 	return func(ctx context.Context, task BenchTask, rc RunContext) TurnTaskOutcome {
+		// buildTurnExecArgs grants the write + sandboxed-shell tool set to EVERY
+		// invocation, so a task MUST run inside an isolated fixture copy — never the
+		// caller's cwd. The grant is unconditional, so the isolation can't be
+		// optional: a fixtureless task would turn an agent with write/shell tools
+		// loose in the caller's working directory (their real repo). Enforce the
+		// invariant instead of trusting it — reject a fixtureless task up front,
+		// before anything is launched, so the grant and the guard can't drift apart.
+		// Every shipped task declares a fixture, so this only fires on a malformed or
+		// newly added task, and it fails loudly rather than silently.
+		if strings.TrimSpace(task.WorkspaceFixture) == "" {
+			return TurnTaskOutcome{Err: errors.New("benchmark task has no workspaceFixture: the run grants write/shell tools and must execute in an isolated fixture copy, not the caller's cwd")}
+		}
 		// Isolate the workspace: copy the fixture into a fresh temp dir so a
 		// mutating task (edit/fix/refactor) can't dirty the shared, checked-in
-		// fixture or bleed into a later iteration of the same task. When no
-		// fixture is configured the agent runs in the caller's cwd as before.
-		if fixture := strings.TrimSpace(task.WorkspaceFixture); fixture != "" {
-			copyDir, parent, cerr := copyFixture(fixture)
-			if cerr != nil {
-				return TurnTaskOutcome{Err: fmt.Errorf("isolate fixture: %w", cerr)}
-			}
-			// Clean the whole unique parent (which owns copyDir) so the
-			// per-invocation scratch dir never leaks.
-			defer os.RemoveAll(parent)
-			task.WorkspaceFixture = copyDir
+		// fixture or bleed into a later iteration of the same task. The guard above
+		// guarantees WorkspaceFixture is set, so this always runs.
+		copyDir, parent, cerr := copyFixture(strings.TrimSpace(task.WorkspaceFixture))
+		if cerr != nil {
+			return TurnTaskOutcome{Err: fmt.Errorf("isolate fixture: %w", cerr)}
 		}
+		// Clean the whole unique parent (which owns copyDir) so the per-invocation
+		// scratch dir never leaks.
+		defer os.RemoveAll(parent)
+		task.WorkspaceFixture = copyDir
 
 		traceFile, err := os.CreateTemp("", "zero-turn-trace-*.ndjson")
 		if err != nil {
@@ -590,10 +682,23 @@ func NewTurnExecRunner(binary string, extraArgs ...string) TurnRunner {
 			}
 		}
 
-		// A nonzero agent exit already decided failure; don't run verification or
-		// mark the task passed.
+		// Decide what a nonzero exit means. We defer to the oracle for exactly ONE
+		// case: an oracle-bearing task that exited INCOMPLETE (exit 4). That is the
+		// completion gate downgrading a run whose edits may be correct but that
+		// couldn't emit a completion signal (e.g. its sandboxed self-verify couldn't
+		// run under --auto member) — the oracle below (the compiled test / grep run
+		// against the actual fixture files) is ground truth, so if the edit really
+		// landed it passes, and if the run abandoned the task the oracle fails it.
+		// Every OTHER nonzero exit stays authoritative: a crash (1), usage error
+		// (2), provider failure (3), or interruption (130) is a genuine failure, so
+		// a partial edit that happens to satisfy the oracle can't launder it into a
+		// pass. A latency-only task has no oracle to defer to, so any nonzero exit
+		// fails it.
 		if outcome.VerifyErr != "" {
-			return outcome
+			if len(task.VerificationCommand) == 0 || exitCode != execExitIncomplete {
+				return outcome
+			}
+			outcome.VerifyErr = ""
 		}
 		// Stamp the compiler-backed oracle and capture the agent's final answer
 		// BEFORE running the verification command. Both write into the fixture
@@ -724,7 +829,23 @@ func stampOracleAndAnswer(task BenchTask, outBuf []byte) error {
 }
 
 func buildTurnExecArgs(task BenchTask, rc RunContext, tracePath string, extraArgs []string) []string {
-	args := []string{"exec", "--output-format", "stream-json", "--trace", tracePath}
+	// --auto member is REQUIRED, not optional: without a permission grant `zero
+	// exec` runs in its default read-only posture, which exposes no write or shell
+	// tools (no edit_file/apply_patch/write_file/exec_command/bash). The mutating
+	// task classes (edit/fix/refactor) then cannot apply any change, so every run
+	// grinds to the turn ceiling or reports a no-tool blocker, and the only tasks
+	// that "pass" are ones whose oracle grep matches the stamped answer text rather
+	// than a real edit. member-auto grants the write + sandboxed-shell tools the
+	// benchmark needs while keeping the workspace/network/destructive safeguards, so
+	// it is preferred over the broader --skip-permissions-unsafe (which also drops
+	// those guards and auto-retries blocked commands unsandboxed). Note the shell it
+	// grants is sandboxed: on a host without sandbox setup the agent's own
+	// self-verification (go test/build) can't run and the turn exits INCOMPLETE even
+	// after a correct edit — the runner treats the stamped oracle, not that exit
+	// code, as ground truth for oracle-bearing tasks (see NewTurnExecRunner), so a
+	// correct edit still passes. Each task also runs in an isolated, throwaway
+	// fixture copy, so the granted tools have nothing outside the task to harm.
+	args := []string{"exec", "--auto", "member", "--output-format", "stream-json", "--trace", tracePath}
 	if model := strings.TrimSpace(rc.Model); model != "" {
 		args = append(args, "--model", model)
 	}
@@ -734,6 +855,9 @@ func buildTurnExecArgs(task BenchTask, rc RunContext, tracePath string, extraArg
 	if rc.SelfCorrect {
 		args = append(args, "--self-correct")
 	}
+	if profile := strings.TrimSpace(rc.ExecProfile); profile != "" {
+		args = append(args, "--exec-profile", profile)
+	}
 	args = append(args, extraArgs...)
 	args = append(args, task.Prompt)
 	return args
@@ -747,7 +871,15 @@ func ResolveBinary(explicit string) (string, error) {
 		if _, err := os.Stat(v); err != nil {
 			return "", fmt.Errorf("trace binary not found: %w", err)
 		}
-		return v, nil
+		// Pin the explicit path to an absolute one: the turn runner sets each
+		// child's cmd.Dir to a per-task fixture copy, so a relative path
+		// (./zero, the Makefile default) that stats fine here would fail to
+		// spawn from inside every fixture dir, silently erroring all tasks.
+		absolute, err := filepath.Abs(v)
+		if err != nil {
+			return "", fmt.Errorf("resolve trace binary path: %w", err)
+		}
+		return absolute, nil
 	}
 	if path, err := exec.LookPath("zero"); err == nil {
 		return path, nil

internal/perfbench/turn_bench_test.go

@@ -4,6 +4,8 @@ import (
 	"bytes"
 	"context"
 	"encoding/json"
+	"errors"
+	"fmt"
 	"os"
 	"path/filepath"
 	"strings"
@@ -530,14 +532,17 @@ func TestRunTurnBenchBuildOnlyMechanism(t *testing.T) {
 	}
 }
 
-// TestTurnSchemaVersion3 pins the schema bump. v3 records the tier
-// reclassification (refactor structural-positive and nav answer-oracles moved
-// into correctnessPassRate), so a v2->v3 cross-version comparison cannot
-// silently misread the jump as a model improvement — exactly the misread the
-// tier system exists to prevent.
-func TestTurnSchemaVersion3(t *testing.T) {
-	if TurnSchemaVersion != 3 {
-		t.Fatalf("TurnSchemaVersion = %d, want 3", TurnSchemaVersion)
+// TestTurnSchemaVersion pins the schema bump so a shape change is a conscious
+// decision. v3 recorded the tier reclassification (refactor structural-positive
+// and nav answer-oracles moved into correctnessPassRate). v4 adds tasksErrored:
+// tasks whose every iteration died before the agent produced a run are now
+// first-class in the report instead of visible only in warnings, so a
+// spawn-broken run cannot print a clean-looking summary or exit 0. v5 is
+// additive only: execProfile stamps the execution profile the run was
+// benchmarked under so profile A/B reports are self-describing.
+func TestTurnSchemaVersion(t *testing.T) {
+	if TurnSchemaVersion != 5 {
+		t.Fatalf("TurnSchemaVersion = %d, want 5", TurnSchemaVersion)
 	}
 }
 
@@ -746,6 +751,92 @@ echo '{"type":"run_end","exitCode":0}'
 	}
 }
 
+// TestOracleAuthoritativeOnIncompleteExit is the companion to the --auto member
+// switch: under member-auto the agent's shell is sandboxed, so on a host without
+// sandbox setup its self-verification (go test/build) can't run and the turn
+// exits INCOMPLETE (exit 4) even after a correct edit. For an ORACLE-bearing task
+// the runner must treat the stamped oracle — not that INCOMPLETE exit — as ground
+// truth: here the stub applies the real edit-01 rename but reports exitCode 4, and
+// the task must still pass because the fixture is correct. This defers ONLY for
+// exit 4; TestNonIncompleteExitStaysAuthoritative pins the other side.
+func TestOracleAuthoritativeOnIncompleteExit(t *testing.T) {
+	task := loadBaselineTask(t, "edit-01")
+	outcome := runTurnStub(t, task, `sed 's/const MaxRetries = 3/const RetryLimit = 3/' main.go > .zero-tmp && mv .zero-tmp main.go
+echo '{"type":"run_end","exitCode":4}'
+`)
+	if outcome.Err != nil {
+		t.Fatalf("incomplete-exit with a correct edit should pass, got harness error: %v", outcome.Err)
+	}
+	if !outcome.Passed {
+		t.Fatalf("a correct edit that exited INCOMPLETE (exit 4) must still pass its oracle: %+v", outcome)
+	}
+	if strings.TrimSpace(outcome.VerifyErr) != "" {
+		t.Fatalf("a passing oracle must clear the exit-code VerifyErr, got %q", outcome.VerifyErr)
+	}
+}
+
+// TestNonIncompleteExitStaysAuthoritative is the guard the oracle-authoritative
+// change MUST NOT weaken: only INCOMPLETE (exit 4) defers to the oracle. A crash
+// (1), provider failure (3), or interruption (130) is a genuine failure, so it
+// stays authoritative even for an oracle-bearing task — otherwise a partial edit
+// that happens to satisfy the oracle would launder a crashed or interrupted run
+// into a pass. Each case applies the CORRECT edit-01 rename (so the oracle WOULD
+// pass) but reports the failing exit code; the task must still fail, and the
+// failure must be attributed to the exit code, not the oracle.
+func TestNonIncompleteExitStaysAuthoritative(t *testing.T) {
+	for _, code := range []int{1, 3, 130} {
+		t.Run(fmt.Sprintf("exit%d", code), func(t *testing.T) {
+			task := loadBaselineTask(t, "edit-01")
+			outcome := runTurnStub(t, task, fmt.Sprintf(`sed 's/const MaxRetries = 3/const RetryLimit = 3/' main.go > .zero-tmp && mv .zero-tmp main.go
+echo '{"type":"run_end","exitCode":%d}'
+`, code))
+			if outcome.Err != nil {
+				t.Fatalf("a nonzero exit should be a task fail, not a harness error: %v", outcome.Err)
+			}
+			if outcome.Passed {
+				t.Fatalf("a run that exited %d must not pass even when the edit satisfies the oracle: %+v", code, outcome)
+			}
+			if want := fmt.Sprintf("exit code %d", code); !strings.Contains(outcome.VerifyErr, want) {
+				t.Fatalf("a non-INCOMPLETE nonzero exit must stay authoritative and surface its code, got VerifyErr=%q", outcome.VerifyErr)
+			}
+		})
+	}
+}
+
+// TestIncompleteExitStillFailsWhenOracleFails pins the OTHER half of the exit-4
+// deferral: deferring to the oracle on INCOMPLETE must still MEAN the oracle is
+// consulted, not a blanket pass. Here the stub does no real work and exits
+// INCOMPLETE (4); the edit-01 rename never happened, so the oracle fails and the
+// task fails. Without this, a regression that turned exit 4 into an unconditional
+// pass would slip through, because its sibling TestOracleAuthoritativeOnIncomplete-
+// Exit applies a CORRECT edit and would pass either way.
+func TestIncompleteExitStillFailsWhenOracleFails(t *testing.T) {
+	task := loadBaselineTask(t, "edit-01")
+	outcome := runTurnStub(t, task, `echo '{"type":"run_end","exitCode":4}'
+`)
+	assertVerifyFailed(t, "incomplete exit with no edit applied", outcome)
+}
+
+// TestNonzeroExitStillFailsLatencyOnly is the guard on the other side of that
+// change: a latency-only task has no oracle to appeal to, so the exit code is
+// the ONLY correctness signal and a nonzero exit must still fail it. Without
+// this, dropping the exit-code gate for oracle tasks could be misread as
+// dropping it everywhere.
+func TestNonzeroExitStillFailsLatencyOnly(t *testing.T) {
+	task := loadBaselineTask(t, "longproc-01")
+	outcome := runTurnStub(t, task, `echo '{"type":"run_end","exitCode":4}'
+`)
+	if outcome.Err != nil {
+		t.Fatalf("latency-only nonzero exit should be a verify fail, not a harness error: %v", outcome.Err)
+	}
+	if outcome.Passed {
+		t.Fatalf("a latency-only task that exited nonzero must not pass: %+v", outcome)
+	}
+	if strings.TrimSpace(outcome.VerifyErr) == "" {
+		t.Fatalf("a latency-only nonzero exit must surface a VerifyErr, got none: %+v", outcome)
+	}
+}
+
 // --- Satisfiable tests: the oracle PASSES the right thing (real fix applied) ---
 
 // TestStampedOraclePassesWhenRefactorHappened proves the refactor-01 oracle is
@@ -1067,3 +1158,238 @@ printf '%s\n' '{"type":"run_end","exitCode":0}'
 `)
 	assertVerifyFailed(t, "nav-08 fmt-only-no-testing", outcome)
 }
+
+func TestResolveBinaryAbsolutizesExplicitPath(t *testing.T) {
+	dir := t.TempDir()
+	name := "zero-probe.exe"
+	if err := os.WriteFile(filepath.Join(dir, name), []byte("x"), 0o755); err != nil {
+		t.Fatal(err)
+	}
+	cwd, err := os.Getwd()
+	if err != nil {
+		t.Fatal(err)
+	}
+	if err := os.Chdir(dir); err != nil {
+		t.Fatal(err)
+	}
+	t.Cleanup(func() { _ = os.Chdir(cwd) })
+
+	resolved, err := ResolveBinary("./" + name)
+	if err != nil {
+		t.Fatalf("ResolveBinary: %v", err)
+	}
+	if !filepath.IsAbs(resolved) {
+		t.Fatalf("ResolveBinary returned relative path %q; the turn runner sets cmd.Dir per task, so a relative binary fails to spawn from every fixture copy", resolved)
+	}
+}
+
+func TestRunTurnBenchCountsErroredTasks(t *testing.T) {
+	set := TaskSet{
+		ID: "errored-suite",
+		Tasks: []BenchTask{
+			{ID: "t1", Class: "nav", Prompt: "p1", VerificationCommand: []string{"true"}},
+			{ID: "t2", Class: "longproc", Prompt: "p2"},
+		},
+	}
+	cfg := TurnBenchConfig{
+		Model:      "fake-model",
+		Iterations: 1,
+		Runner: func(context.Context, BenchTask, RunContext) TurnTaskOutcome {
+			return TurnTaskOutcome{Err: errors.New("fork/exec ./zero: file does not exist")}
+		},
+		Now: func() time.Time { return time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC) },
+	}
+	result, err := RunTurnBench(context.Background(), set, cfg)
+	if err != nil {
+		t.Fatalf("RunTurnBench: %v", err)
+	}
+	if result.TasksErrored != 2 {
+		t.Fatalf("TasksErrored=%d, want 2 (every iteration of both tasks errored)", result.TasksErrored)
+	}
+	if len(result.Warnings) != 2 {
+		t.Fatalf("warnings=%d, want 2 run warnings", len(result.Warnings))
+	}
+	summary := FormatTurnBenchSummary(result)
+	if !strings.Contains(summary, "ERRORED: 2 task(s)") {
+		t.Fatalf("summary does not surface the errored tasks:\n%s", summary)
+	}
+	if !strings.Contains(summary, "fork/exec ./zero") {
+		t.Fatalf("summary does not echo the underlying run error:\n%s", summary)
+	}
+}
+
+func TestRunTurnBenchPartialErrorStillCounts(t *testing.T) {
+	set := TaskSet{
+		ID: "partial-suite",
+		Tasks: []BenchTask{
+			{ID: "ok", Class: "nav", Prompt: "p", VerificationCommand: []string{"true"}},
+			{ID: "dead", Class: "nav", Prompt: "p", VerificationCommand: []string{"true"}},
+		},
+	}
+	canned := map[string]*trace.TurnTrace{"ok": cannedTrace(100, 10, 1000)}
+	inner := fakeTurnRunner(canned)
+	cfg := TurnBenchConfig{
+		Model:      "fake-model",
+		Iterations: 1,
+		Runner: func(ctx context.Context, task BenchTask, rc RunContext) TurnTaskOutcome {
+			if task.ID == "dead" {
+				return TurnTaskOutcome{Err: errors.New("spawn failed")}
+			}
+			return inner(ctx, task, rc)
+		},
+		Now: func() time.Time { return time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC) },
+	}
+	result, err := RunTurnBench(context.Background(), set, cfg)
+	if err != nil {
+		t.Fatalf("RunTurnBench: %v", err)
+	}
+	if result.TasksErrored != 1 {
+		t.Fatalf("TasksErrored=%d, want 1", result.TasksErrored)
+	}
+	if result.TasksAttempted != 2 {
+		t.Fatalf("TasksAttempted=%d, want 2", result.TasksAttempted)
+	}
+}
+
+// The exec-profile passthrough: the profile must reach every task's zero exec
+// invocation as --exec-profile (with the prompt staying last) and stay out of
+// the args entirely when unset.
+func TestBuildTurnExecArgsIncludesExecProfile(t *testing.T) {
+	task := BenchTask{ID: "t", Prompt: "do the thing"}
+	args := buildTurnExecArgs(task, RunContext{Model: "m", ExecProfile: "fast"}, "trace.ndjson", nil)
+	found := false
+	for i := 0; i < len(args)-1; i++ {
+		if args[i] == "--exec-profile" && args[i+1] == "fast" {
+			found = true
+			break
+		}
+	}
+	if !found {
+		t.Fatalf("args must carry --exec-profile fast, got %v", args)
+	}
+	if args[len(args)-1] != "do the thing" {
+		t.Fatalf("prompt must stay the last argument, got %v", args)
+	}
+
+	args = buildTurnExecArgs(task, RunContext{Model: "m"}, "trace.ndjson", nil)
+	for _, arg := range args {
+		if arg == "--exec-profile" {
+			t.Fatalf("no profile configured, but args carry --exec-profile: %v", args)
+		}
+	}
+}
+
+// Every benchmark invocation MUST grant the write/shell tool set via --auto
+// member. Without a permission grant the agent runs read-only and cannot apply
+// any edit, so the mutating classes measure nothing but oracle/answer
+// contamination. member-auto is the RIGHT grant: it exposes the write +
+// sandboxed-shell tools the benchmark needs while keeping the workspace/network/
+// destructive safeguards, so the args must NOT reach for the broader
+// --skip-permissions-unsafe (which drops those guards). This is a correctness
+// contract, not a preference.
+func TestBuildTurnExecArgsGrantsWriteTools(t *testing.T) {
+	args := buildTurnExecArgs(BenchTask{ID: "t", Prompt: "edit the file"}, RunContext{Model: "m"}, "trace.ndjson", nil)
+	autoMember := false
+	for i := 0; i < len(args)-1; i++ {
+		if args[i] == "--auto" && args[i+1] == "member" {
+			autoMember = true
+			break
+		}
+	}
+	if !autoMember {
+		t.Fatalf("benchmark exec args must include --auto member so the agent can apply edits, got %v", args)
+	}
+	for _, arg := range args {
+		if arg == "--skip-permissions-unsafe" {
+			t.Fatalf("benchmark exec args must NOT use --skip-permissions-unsafe (member-auto keeps the sandbox/network guards), got %v", args)
+		}
+	}
+	if args[len(args)-1] != "edit the file" {
+		t.Fatalf("prompt must stay the last argument, got %v", args)
+	}
+}
+
+// TestTurnRunnerRejectsFixturelessTask is the other half of the write-tools
+// contract: because buildTurnExecArgs grants the write + sandboxed-shell tool
+// set to every invocation, a task with no fixture would run an agent with those
+// tools in the caller's cwd. The runner must reject such a task BEFORE launching
+// zero exec. Passing a binary path that does not exist proves the rejection is
+// pre-launch — the outcome carries the fixture error, not a spawn error — so this
+// runs on every OS (it never reaches the POSIX-only exec stub).
+func TestTurnRunnerRejectsFixturelessTask(t *testing.T) {
+	task := BenchTask{ID: "no-fixture", Prompt: "do a thing"} // no WorkspaceFixture set
+	outcome := NewTurnExecRunner(filepath.Join(t.TempDir(), "does-not-exist-zero"))(context.Background(), task, RunContext{Model: "m"})
+	if outcome.Err == nil {
+		t.Fatalf("a fixtureless task must be rejected before launch, got %+v", outcome)
+	}
+	if !strings.Contains(outcome.Err.Error(), "workspaceFixture") {
+		t.Fatalf("rejection must name the missing workspaceFixture, got: %v", outcome.Err)
+	}
+	if outcome.Passed {
+		t.Fatalf("a rejected task must not pass, got %+v", outcome)
+	}
+}
+
+// The configured profile must reach the runner's RunContext and be stamped
+// into the result, so a profile A/B report is self-describing. The boundary
+// canonicalizes (case/whitespace) so equivalent postures always carry the same
+// label, and rejects unknown names so a direct library caller cannot slip an
+// unvalidated profile into a report the way the CLI (parse-time check) cannot.
+func TestRunTurnBenchStampsExecProfile(t *testing.T) {
+	set := TaskSet{
+		ID:    "profile-suite",
+		Tasks: []BenchTask{{ID: "t1", Class: "longproc", Prompt: "p"}},
+	}
+	var gotProfile string
+	cfg := TurnBenchConfig{
+		Model:       "fake-model",
+		ExecProfile: " FAST ",
+		Iterations:  1,
+		Runner: func(_ context.Context, _ BenchTask, rc RunContext) TurnTaskOutcome {
+			gotProfile = rc.ExecProfile
+			return TurnTaskOutcome{Passed: true, WallMs: 10, Trace: cannedTrace(10, 1, 100)}
+		},
+		Now: func() time.Time { return time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC) },
+	}
+	result, err := RunTurnBench(context.Background(), set, cfg)
+	if err != nil {
+		t.Fatalf("RunTurnBench: %v", err)
+	}
+	if gotProfile != "fast" {
+		t.Fatalf("runner RunContext.ExecProfile = %q, want the canonical fast", gotProfile)
+	}
+	if result.ExecProfile != "fast" {
+		t.Fatalf("result.ExecProfile = %q, want the canonical fast", result.ExecProfile)
+	}
+	raw, err := json.Marshal(result)
+	if err != nil {
+		t.Fatalf("marshal result: %v", err)
+	}
+	if !strings.Contains(string(raw), `"execProfile":"fast"`) {
+		t.Fatal("published JSON must carry execProfile")
+	}
+}
+
+func TestRunTurnBenchRejectsUnknownExecProfile(t *testing.T) {
+	set := TaskSet{
+		ID:    "profile-suite",
+		Tasks: []BenchTask{{ID: "t1", Class: "longproc", Prompt: "p"}},
+	}
+	ran := false
+	cfg := TurnBenchConfig{
+		Model:       "fake-model",
+		ExecProfile: "blanced",
+		Iterations:  1,
+		Runner: func(context.Context, BenchTask, RunContext) TurnTaskOutcome {
+			ran = true
+			return TurnTaskOutcome{Passed: true, WallMs: 10, Trace: cannedTrace(10, 1, 100)}
+		},
+	}
+	_, err := RunTurnBench(context.Background(), set, cfg)
+	if err == nil || !strings.Contains(err.Error(), "unknown execution profile") {
+		t.Fatalf("err = %v, want an unknown-profile rejection", err)
+	}
+	if ran {
+		t.Fatal("nothing may run under an unknown profile")
+	}
+}

internal/providers/factory.go

@@ -108,6 +108,57 @@ func New(profile config.ProviderProfile, options Options) (zeroruntime.Provider,
 	}
 }
 
+// NewTurnSessionProvider builds the default TurnSessionProvider for a resolved
+// profile: it constructs the provider exactly as New does, then wraps it with a
+// ProviderCapabilities projection computed from the same model-registry entry
+// New already resolves. The default session's Stream is the provider's
+// StreamCompletion, so runtime behavior is identical to using New directly.
+func NewTurnSessionProvider(profile config.ProviderProfile, options Options) (zeroruntime.TurnSessionProvider, error) {
+	provider, err := New(profile, options)
+	if err != nil {
+		return nil, err
+	}
+	caps, err := resolveCapabilities(profile, options)
+	if err != nil {
+		return nil, err
+	}
+	return zeroruntime.NewProviderTurnSessionProvider(provider, caps), nil
+}
+
+// resolveCapabilities projects the resolved profile's model-registry entry into
+// the flat zeroruntime.ProviderCapabilities (zeroruntime cannot reference
+// modelregistry types — modelregistry imports zeroruntime, so a typed field
+// would form an import cycle). A model absent from the registry yields only the
+// resolved API model id and max-output tokens; the rest stays zero (unknown).
+func resolveCapabilities(profile config.ProviderProfile, options Options) (zeroruntime.ProviderCapabilities, error) {
+	resolved, err := resolveProfile(profile, options)
+	if err != nil {
+		return zeroruntime.ProviderCapabilities{}, err
+	}
+	caps := zeroruntime.ProviderCapabilities{
+		Model:           resolved.apiModel,
+		MaxOutputTokens: resolved.maxOutputTokens,
+	}
+	registry, err := defaultRegistry(options.ModelRegistry)
+	if err != nil {
+		return zeroruntime.ProviderCapabilities{}, err
+	}
+	if entry, ok := registry.Get(strings.TrimSpace(profile.Model)); ok {
+		caps.ContextWindow = entry.ContextLimits.ContextWindow
+		caps.SupportsVision = entry.Supports(modelregistry.ModelCapabilityVision)
+		caps.SupportsReasoning = entry.Supports(modelregistry.ModelCapabilityReasoning)
+		caps.SupportsPromptCache = entry.Supports(modelregistry.ModelCapabilityPromptCache)
+		// Read efforts through Registry.ReasoningEfforts (not the raw entry) so
+		// the projection matches what the /effort picker and the run-time
+		// resolver advertise — including the name-based fallback for catalog
+		// entries that enumerate no efforts of their own.
+		for _, effort := range registry.ReasoningEfforts(entry.ID) {
+			caps.ReasoningEfforts = append(caps.ReasoningEfforts, string(effort))
+		}
+	}
+	return caps, nil
+}
+
 func parseThinkTagsForProfile(profile config.ProviderProfile, resolved resolvedProfile) bool {
 	if profile.ParseThinkTags != nil {
 		return *profile.ParseThinkTags

internal/providers/factory_turn_session_test.go

@@ -0,0 +1,194 @@
+package providers
+
+import (
+	"context"
+	"errors"
+	"testing"
+
+	"github.com/Gitlawb/zero/internal/config"
+	"github.com/Gitlawb/zero/internal/modelregistry"
+	"github.com/Gitlawb/zero/internal/zeroruntime"
+)
+
+// TestNewTurnSessionProviderForEveryKind asserts every current provider kind
+// wraps into the default TurnSessionProvider: construction succeeds, a session
+// opens, and Compact reports unsupported (the default adapter contract).
+func TestNewTurnSessionProviderForEveryKind(t *testing.T) {
+	kinds := []config.ProviderKind{
+		config.ProviderKindOpenAI,
+		config.ProviderKindOpenAICompatible,
+		config.ProviderKindAnthropic,
+		config.ProviderKindAnthropicCompat,
+		config.ProviderKindGoogle,
+	}
+	for _, kind := range kinds {
+		t.Run(string(kind), func(t *testing.T) {
+			tsp, err := NewTurnSessionProvider(config.ProviderProfile{
+				Name:         "pr7-" + string(kind),
+				ProviderKind: kind,
+				BaseURL:      "https://provider.example/v1",
+				APIKey:       "sk-turn-session-test",
+				// A model absent from the registry: resolution falls back to the
+				// raw id, so this test stays hermetic across catalog changes.
+				Model: "pr7-unregistered-model",
+			}, Options{UserAgent: "zero-turn-session-test"})
+			if err != nil {
+				t.Fatalf("NewTurnSessionProvider(%s): %v", kind, err)
+			}
+			session, err := tsp.OpenTurnSession(context.Background())
+			if err != nil {
+				t.Fatalf("OpenTurnSession(%s): %v", kind, err)
+			}
+			defer func() {
+				if closeErr := session.Close(); closeErr != nil {
+					t.Fatalf("Close(%s): %v", kind, closeErr)
+				}
+			}()
+			if err := session.Prewarm(context.Background()); err != nil {
+				t.Fatalf("Prewarm(%s): %v", kind, err)
+			}
+			if _, compactErr := session.Compact(context.Background(), zeroruntime.CompletionRequest{}); !errors.Is(compactErr, zeroruntime.ErrCompactionUnsupported) {
+				t.Fatalf("Compact(%s) = %v, want ErrCompactionUnsupported", kind, compactErr)
+			}
+			caps := tsp.Capabilities()
+			if caps.Model != "pr7-unregistered-model" {
+				t.Fatalf("Capabilities(%s).Model = %q, want the resolved api model", kind, caps.Model)
+			}
+			if caps.NativeCompaction {
+				t.Fatalf("Capabilities(%s).NativeCompaction = true, want false for the default adapter", kind)
+			}
+		})
+	}
+}
+
+// TestNewTurnSessionProviderProjectsRegistryCapabilities asserts the factory
+// projects the model-registry entry (context limits, capability flags,
+// reasoning efforts) into the flat ProviderCapabilities.
+func TestNewTurnSessionProviderProjectsRegistryCapabilities(t *testing.T) {
+	registry, err := modelregistry.NewRegistry([]modelregistry.ModelEntry{{
+		ID:          "pr7-caps-model",
+		DisplayName: "PR7 Capability Probe",
+		APIModel:    "pr7-caps-api-model",
+		Provider:    modelregistry.ProviderOpenAI,
+		ContextLimits: modelregistry.ContextLimits{
+			ContextWindow:   200_000,
+			MaxOutputTokens: 64_000,
+		},
+		ReasoningEfforts: []modelregistry.ReasoningEffort{
+			modelregistry.ReasoningEffortLow,
+			modelregistry.ReasoningEffortMedium,
+			modelregistry.ReasoningEffortHigh,
+		},
+		Capabilities: modelregistry.ModelCapabilities{
+			modelregistry.ModelCapabilityChat,
+			modelregistry.ModelCapabilityVision,
+			modelregistry.ModelCapabilityReasoning,
+			modelregistry.ModelCapabilityPromptCache,
+		},
+		Cost: modelregistry.ModelCost{
+			Currency:           "USD",
+			Unit:               "per_1m_tokens",
+			InputPerMillion:    1,
+			OutputPerMillion:   2,
+			Source:             "https://provider.example/pricing (test fixture)",
+			SourceLastVerified: "2026-07-18",
+		},
+		Status:  modelregistry.ModelStatusActive,
+		Aliases: []string{"pr7-caps"},
+	}})
+	if err != nil {
+		t.Fatalf("NewRegistry: %v", err)
+	}
+
+	tsp, err := NewTurnSessionProvider(config.ProviderProfile{
+		Name:         "pr7-caps",
+		ProviderKind: config.ProviderKindOpenAI,
+		BaseURL:      "https://provider.example/v1",
+		APIKey:       "sk-turn-session-test",
+		Model:        "pr7-caps-model",
+	}, Options{
+		UserAgent:     "zero-turn-session-test",
+		ModelRegistry: &registry,
+	})
+	if err != nil {
+		t.Fatalf("NewTurnSessionProvider: %v", err)
+	}
+
+	caps := tsp.Capabilities()
+	if caps.Model != "pr7-caps-api-model" {
+		t.Fatalf("Model = %q, want the registry APIModel", caps.Model)
+	}
+	if caps.ContextWindow != 200_000 || caps.MaxOutputTokens != 64_000 {
+		t.Fatalf("limits = (%d, %d), want (200000, 64000)", caps.ContextWindow, caps.MaxOutputTokens)
+	}
+	if !caps.SupportsVision || !caps.SupportsReasoning || !caps.SupportsPromptCache {
+		t.Fatalf("capability flags = %+v, want vision/reasoning/prompt-cache all true", caps)
+	}
+	if len(caps.ReasoningEfforts) != 3 || caps.ReasoningEfforts[0] != "low" || caps.ReasoningEfforts[2] != "high" {
+		t.Fatalf("ReasoningEfforts = %v, want [low medium high]", caps.ReasoningEfforts)
+	}
+	if caps.NativeCompaction {
+		t.Fatal("NativeCompaction = true, want false for the default adapter")
+	}
+}
+
+// TestNewTurnSessionProviderUsesEffectiveReasoningEfforts asserts the
+// projection reads efforts through Registry.ReasoningEfforts, so a catalog
+// entry that enumerates no efforts of its own still reports the name-inferred
+// effective tiers the /effort picker and run-time resolver advertise.
+func TestNewTurnSessionProviderUsesEffectiveReasoningEfforts(t *testing.T) {
+	registry, err := modelregistry.NewRegistry([]modelregistry.ModelEntry{{
+		// A gpt-5-family id with NO ReasoningEfforts listed: the effective
+		// efforts come from name inference, differing from the raw entry.
+		ID:          "gpt-5-pr7-probe",
+		DisplayName: "PR7 Effective Efforts Probe",
+		APIModel:    "gpt-5-pr7-probe-api",
+		Provider:    modelregistry.ProviderOpenAI,
+		ContextLimits: modelregistry.ContextLimits{
+			ContextWindow:   400_000,
+			MaxOutputTokens: 128_000,
+		},
+		Capabilities: modelregistry.ModelCapabilities{
+			modelregistry.ModelCapabilityChat,
+			modelregistry.ModelCapabilityReasoning,
+		},
+		Cost: modelregistry.ModelCost{
+			Currency:           "USD",
+			Unit:               "per_1m_tokens",
+			InputPerMillion:    1,
+			OutputPerMillion:   2,
+			Source:             "https://provider.example/pricing (test fixture)",
+			SourceLastVerified: "2026-07-18",
+		},
+		Status:  modelregistry.ModelStatusActive,
+		Aliases: []string{"pr7-effective-efforts"},
+	}})
+	if err != nil {
+		t.Fatalf("NewRegistry: %v", err)
+	}
+
+	tsp, err := NewTurnSessionProvider(config.ProviderProfile{
+		Name:         "pr7-effective",
+		ProviderKind: config.ProviderKindOpenAI,
+		BaseURL:      "https://provider.example/v1",
+		APIKey:       "sk-turn-session-test",
+		Model:        "gpt-5-pr7-probe",
+	}, Options{
+		UserAgent:     "zero-turn-session-test",
+		ModelRegistry: &registry,
+	})
+	if err != nil {
+		t.Fatalf("NewTurnSessionProvider: %v", err)
+	}
+
+	got := tsp.Capabilities().ReasoningEfforts
+	want := []string{"minimal", "low", "medium", "high"}
+	if len(got) != len(want) {
+		t.Fatalf("ReasoningEfforts = %v, want %v (name-inferred fallback)", got, want)
+	}
+	for i := range want {
+		if got[i] != want[i] {
+			t.Fatalf("ReasoningEfforts[%d] = %q, want %q (full: %v)", i, got[i], want[i], got)
+		}
+	}
+}

internal/providers/openai/provider_test.go

@@ -5,6 +5,7 @@ import (
 	"encoding/json"
 	"net/http"
 	"net/http/httptest"
+	"reflect"
 	"strings"
 	"testing"
 	"time"
@@ -1375,3 +1376,61 @@ func TestOpenAIRequestPromptCacheKey(t *testing.T) {
 		}
 	}
 }
+
+func TestOpenAIRequestPreservesCacheablePrefixAcrossTurns(t *testing.T) {
+	provider, err := New(Options{Model: "gpt-test"})
+	if err != nil {
+		t.Fatalf("New returned error: %v", err)
+	}
+	tools := []zeroruntime.ToolDefinition{{
+		Name:        "read_file",
+		Description: "Read a file.",
+		Parameters: map[string]any{
+			"type": "object",
+			"properties": map[string]any{
+				"path": map[string]any{"type": "string"},
+			},
+		},
+	}}
+	firstMessages := []zeroruntime.Message{
+		{Role: zeroruntime.MessageRoleSystem, Content: "stable system prompt"},
+		{Role: zeroruntime.MessageRoleUser, Content: "first turn"},
+	}
+	secondMessages := append([]zeroruntime.Message(nil), firstMessages...)
+	secondMessages = append(secondMessages,
+		zeroruntime.Message{Role: zeroruntime.MessageRoleAssistant, Content: "first response"},
+		zeroruntime.Message{Role: zeroruntime.MessageRoleUser, Content: "second turn"},
+	)
+
+	marshalBody := func(messages []zeroruntime.Message) map[string]any {
+		t.Helper()
+		mapped := provider.openAIRequest(zeroruntime.CompletionRequest{
+			Messages:       messages,
+			Tools:          tools,
+			PromptCacheKey: "session-stable-prefix",
+		})
+		data, marshalErr := json.Marshal(mapped)
+		if marshalErr != nil {
+			t.Fatalf("marshal request: %v", marshalErr)
+		}
+		var body map[string]any
+		if unmarshalErr := json.Unmarshal(data, &body); unmarshalErr != nil {
+			t.Fatalf("unmarshal request: %v", unmarshalErr)
+		}
+		return body
+	}
+
+	first := marshalBody(firstMessages)
+	second := marshalBody(secondMessages)
+	firstWireMessages := first["messages"].([]any)
+	secondWireMessages := second["messages"].([]any)
+	if !reflect.DeepEqual(secondWireMessages[:len(firstWireMessages)], firstWireMessages) {
+		t.Fatalf("first wire messages must be an exact prefix of the second:\nfirst=%#v\nsecond=%#v", firstWireMessages, secondWireMessages)
+	}
+	if !reflect.DeepEqual(first["tools"], second["tools"]) {
+		t.Fatalf("wire tool definitions drifted:\nfirst=%#v\nsecond=%#v", first["tools"], second["tools"])
+	}
+	if first["prompt_cache_key"] != second["prompt_cache_key"] || first["prompt_cache_key"] != "session-stable-prefix" {
+		t.Fatalf("wire prompt cache key must remain stable: first=%#v second=%#v", first["prompt_cache_key"], second["prompt_cache_key"])
+	}
+}

internal/providers/openai/session.go

@@ -0,0 +1,205 @@
+package openai
+
+import (
+	"context"
+	"crypto/sha256"
+	"encoding/hex"
+	"encoding/json"
+	"fmt"
+	"net/http"
+	"strings"
+	"sync"
+	"time"
+
+	"github.com/Gitlawb/zero/internal/trace"
+	"github.com/Gitlawb/zero/internal/zeroruntime"
+)
+
+// prewarmTimeout bounds the best-effort prewarm probe. The probe runs in the
+// background, so this cap only limits how long the goroutine may linger — it
+// can never delay the first turn.
+const prewarmTimeout = 3 * time.Second
+
+// NewTurnSessionProvider wraps an already-constructed *Provider in the
+// optimized OpenAI turn session: a best-effort connection prewarm at run start
+// plus request-prefix stability telemetry. Stream is Provider.StreamCompletion
+// verbatim, so runtime request behavior is identical to the default adapter —
+// the session adds observation and pool priming, never a different request.
+func NewTurnSessionProvider(provider *Provider, caps zeroruntime.ProviderCapabilities) zeroruntime.TurnSessionProvider {
+	return &turnSessionProvider{provider: provider, caps: caps}
+}
+
+type turnSessionProvider struct {
+	provider *Provider
+	caps     zeroruntime.ProviderCapabilities
+}
+
+func (p *turnSessionProvider) OpenTurnSession(context.Context) (zeroruntime.TurnSession, error) {
+	return &turnSession{provider: p.provider}, nil
+}
+
+func (p *turnSessionProvider) Capabilities() zeroruntime.ProviderCapabilities {
+	return p.caps
+}
+
+// turnSession is one run's optimized OpenAI session. The agent loop serializes
+// all provider I/O through its session shim, so fields need no mutex. The
+// fingerprint counters exist so a future stateful-reuse session (Responses API)
+// inherits a proven prefix-stability detector; on chat completions there is no
+// server-side response state to invalidate, so drift is telemetry only.
+type turnSession struct {
+	provider        *Provider
+	lastFingerprint string
+	prewarmOnce     sync.Once
+	// prewarmDone closes when the background probe settles; tests wait on it.
+	prewarmDone chan struct{}
+}
+
+// Prewarm launches one bounded, unauthenticated HEAD probe to the provider's
+// base URL in the background so the TCP+TLS handshake lands in the shared
+// connection pool while the loop assembles the first prompt. One attempt, no
+// retries, no bearer token — the goal is the handshake, not a 2xx (a 401/404/
+// 405 response still primes the pool). Always returns nil: prewarm is advisory
+// by contract and never required for correctness.
+//
+// The probe is issued directly through the client (NOT providerio.SendWithRetry)
+// so it stamps ONLY the provider_prewarm span — a nested provider_connect stamp
+// would contaminate the exact A/B metric this feature is measured by.
+//
+// When the client's transport cannot retain idle connections (macOS disables
+// keep-alives on the shared transport because degraded pooled connections are
+// indistinguishable from backend slowness there), the probe is skipped entirely
+// rather than spending a request that can never be reused. The pool's 30s idle
+// timeout bounds the warm window; a slower start simply falls back to a cold
+// dial, never worse than today.
+func (s *turnSession) Prewarm(ctx context.Context) error {
+	s.prewarmOnce.Do(func() {
+		done := make(chan struct{})
+		s.prewarmDone = done
+		provider := s.provider
+		if !transportRetainsIdleConns(provider.httpClient) {
+			close(done)
+			return
+		}
+		recorder := trace.FromContext(ctx)
+		go func() {
+			defer close(done)
+			probeCtx, cancel := context.WithTimeout(ctx, prewarmTimeout)
+			defer cancel()
+			span := recorder.Span(trace.SpanProviderPrewarm)
+			defer span.End()
+			recorder.Counter(trace.CounterPrewarmAttempts, 1)
+			request, err := http.NewRequestWithContext(probeCtx, http.MethodHead, provider.baseURL, nil)
+			if err != nil {
+				return
+			}
+			if provider.userAgent != "" {
+				request.Header.Set("User-Agent", provider.userAgent)
+			}
+			response, err := provider.httpClient.Do(request)
+			if err == nil {
+				_ = response.Body.Close()
+			}
+		}()
+	})
+	return nil
+}
+
+// transportRetainsIdleConns reports whether a warmed connection can outlive the
+// probe. A *http.Transport with DisableKeepAlives (the shared transport on
+// macOS) closes every connection after its request, so a prewarm there is pure
+// overhead. A nil transport is net/http's default (keep-alives on); an unknown
+// custom transport is assumed reusable — the probe stays a harmless best-effort.
+func transportRetainsIdleConns(client *http.Client) bool {
+	if client == nil {
+		return false
+	}
+	if transport, ok := client.Transport.(*http.Transport); ok {
+		return !transport.DisableKeepAlives
+	}
+	return true
+}
+
+func (s *turnSession) Stream(ctx context.Context, request zeroruntime.CompletionRequest) (<-chan zeroruntime.StreamEvent, error) {
+	s.observePrefix(ctx, request)
+	return s.provider.StreamCompletion(ctx, request)
+}
+
+func (s *turnSession) Compact(context.Context, zeroruntime.CompletionRequest) ([]zeroruntime.Message, error) {
+	return nil, zeroruntime.ErrCompactionUnsupported
+}
+
+// Close is an idempotent no-op: the shared transport pool owns the connections.
+func (s *turnSession) Close() error { return nil }
+
+// observePrefix tracks whether the request-prefix parameters stayed stable
+// between this session's streams. The first stream seeds the fingerprint and
+// counts as neither stable nor drift.
+func (s *turnSession) observePrefix(ctx context.Context, request zeroruntime.CompletionRequest) {
+	fingerprint := s.computeFingerprint(request)
+	if s.lastFingerprint == "" {
+		s.lastFingerprint = fingerprint
+		return
+	}
+	recorder := trace.FromContext(ctx)
+	if fingerprint == s.lastFingerprint {
+		recorder.Counter(trace.CounterPrefixStable, 1)
+		return
+	}
+	recorder.Counter(trace.CounterPrefixDrift, 1)
+	s.lastFingerprint = fingerprint
+}
+
+// computeFingerprint digests the wire-affecting request parameters: the
+// session's model and max-tokens (fixed at construction — CompletionRequest
+// carries no model field), the normalized reasoning effort as it would appear
+// on the wire, the prompt-cache key (constant within a session, and part of
+// the request the provider serializes), and an order-preserving digest of the
+// advertised tools (request serialization preserves tool order, so a reorder
+// changes the wire bytes and counts as drift). Messages are excluded — they
+// grow every turn; this fingerprints request parameters, not conversation
+// content.
+//
+// SCOPE: this is per-provider request-parameter TELEMETRY, not a complete
+// compatibility fingerprint. It deliberately does not incorporate the prompt
+// prefix hashes (base instructions / project context / skills — the
+// complete_prefix signal the agent loop already emits per turn). Any future
+// stateful response reuse MUST combine both signals plus the remaining
+// wire-affecting fields before reusing provider-side state; these counters
+// alone are not sufficient compatibility protection.
+func (s *turnSession) computeFingerprint(request zeroruntime.CompletionRequest) string {
+	var builder strings.Builder
+	builder.WriteString(s.provider.model)
+	builder.WriteByte('|')
+	fmt.Fprintf(&builder, "%d", s.provider.maxTokens)
+	builder.WriteByte('|')
+	builder.WriteString(openAIReasoningEffort(request.ReasoningEffort))
+	builder.WriteByte('|')
+	builder.WriteString(request.PromptCacheKey)
+	builder.WriteByte('|')
+	builder.WriteString(wireToolsDigest(request.Tools))
+	sum := sha256.Sum256([]byte(builder.String()))
+	return hex.EncodeToString(sum[:])
+}
+
+// wireToolsDigest renders the tool set in WIRE ORDER: each tool as name +
+// description + its JSON-marshaled parameters (encoding/json sorts map keys,
+// so each render is stable). Order is preserved deliberately — the OpenAI
+// request serializes tools in the order given, so a reorder changes the
+// request bytes and must count as drift. Descriptions are model-visible
+// request bytes too. A schema that fails to marshal is recorded under a
+// stable sentinel — this digest feeds telemetry counters, so cruder handling
+// of that pathological case is acceptable.
+func wireToolsDigest(tools []zeroruntime.ToolDefinition) string {
+	rendered := make([]string, 0, len(tools))
+	for _, tool := range tools {
+		schema, err := json.Marshal(tool.Parameters)
+		if err != nil {
+			rendered = append(rendered, tool.Name+"\n"+tool.Description+"\n__non_json:"+tool.Name)
+			continue
+		}
+		rendered = append(rendered, tool.Name+"\n"+tool.Description+"\n"+string(schema))
+	}
+	sum := sha256.Sum256([]byte(strings.Join(rendered, "\x00")))
+	return hex.EncodeToString(sum[:])
+}

internal/providers/openai/session_test.go

@@ -0,0 +1,387 @@
+package openai
+
+import (
+	"context"
+	"errors"
+	"net/http"
+	"net/http/httptest"
+	"sync/atomic"
+	"testing"
+	"time"
+
+	"github.com/Gitlawb/zero/internal/trace"
+	"github.com/Gitlawb/zero/internal/zeroruntime"
+)
+
+// keepAliveClient returns a client whose transport retains idle connections on
+// EVERY platform. The shared default transport disables keep-alives on macOS,
+// where the session correctly skips the probe — tests that assert the probe
+// fires must therefore pin a reusable transport to stay platform-deterministic.
+func keepAliveClient() *http.Client {
+	transport := http.DefaultTransport.(*http.Transport).Clone()
+	transport.DisableKeepAlives = false
+	return &http.Client{Transport: transport}
+}
+
+// newPrewarmTestProvider builds a provider whose transport always allows the
+// prewarm probe, regardless of host platform.
+func newPrewarmTestProvider(t *testing.T, handler http.HandlerFunc) *Provider {
+	t.Helper()
+	return newTestProviderWithOptions(t, Options{APIKey: "sk-test", HTTPClient: keepAliveClient()}, handler)
+}
+
+func waitPrewarmDone(t *testing.T, session *turnSession) {
+	t.Helper()
+	if session.prewarmDone == nil {
+		t.Fatal("prewarm was never launched")
+	}
+	select {
+	case <-session.prewarmDone:
+	case <-time.After(10 * time.Second):
+		t.Fatal("prewarm probe did not settle in time")
+	}
+}
+
+func openOptimizedSession(t *testing.T, provider *Provider) *turnSession {
+	t.Helper()
+	session, err := NewTurnSessionProvider(provider, zeroruntime.ProviderCapabilities{}).OpenTurnSession(context.Background())
+	if err != nil {
+		t.Fatalf("OpenTurnSession: %v", err)
+	}
+	return session.(*turnSession)
+}
+
+func TestTurnSessionPrewarmSendsSingleHEAD(t *testing.T) {
+	var requests atomic.Int64
+	var lastMethod atomic.Value
+	provider := newPrewarmTestProvider(t, func(w http.ResponseWriter, r *http.Request) {
+		requests.Add(1)
+		lastMethod.Store(r.Method)
+		w.WriteHeader(http.StatusMethodNotAllowed)
+	})
+
+	session := openOptimizedSession(t, provider)
+	if err := session.Prewarm(context.Background()); err != nil {
+		t.Fatalf("Prewarm: %v", err)
+	}
+	waitPrewarmDone(t, session)
+
+	if got := requests.Load(); got != 1 {
+		t.Fatalf("prewarm sent %d requests, want exactly 1 (no retries even on 405)", got)
+	}
+	if method, _ := lastMethod.Load().(string); method != http.MethodHead {
+		t.Fatalf("prewarm method = %q, want HEAD", method)
+	}
+
+	// A second Prewarm (e.g. after a mid-run swap re-open) must not re-probe.
+	if err := session.Prewarm(context.Background()); err != nil {
+		t.Fatalf("second Prewarm: %v", err)
+	}
+	waitPrewarmDone(t, session)
+	if got := requests.Load(); got != 1 {
+		t.Fatalf("second Prewarm re-probed: %d requests, want 1", got)
+	}
+}
+
+func TestTurnSessionPrewarmNonFatalOnConnectError(t *testing.T) {
+	// A server that is immediately closed leaves a port that refuses connections.
+	server := httptest.NewServer(http.HandlerFunc(func(http.ResponseWriter, *http.Request) {}))
+	deadURL := server.URL
+	server.Close()
+
+	provider, err := New(Options{APIKey: "sk-test", BaseURL: deadURL, Model: "gpt-test", HTTPClient: keepAliveClient()})
+	if err != nil {
+		t.Fatalf("New: %v", err)
+	}
+	session := openOptimizedSession(t, provider)
+	if err := session.Prewarm(context.Background()); err != nil {
+		t.Fatalf("Prewarm must be advisory even when the probe cannot connect, got %v", err)
+	}
+	waitPrewarmDone(t, session)
+}
+
+func TestTurnSessionPrewarmTraceStamps(t *testing.T) {
+	provider := newPrewarmTestProvider(t, func(w http.ResponseWriter, _ *http.Request) {
+		w.WriteHeader(http.StatusMethodNotAllowed)
+	})
+	recorder := trace.NewRecorder("session-test", "run-1", "test")
+	recorder.Start()
+	ctx := trace.WithContext(context.Background(), recorder)
+
+	session := openOptimizedSession(t, provider)
+	if err := session.Prewarm(ctx); err != nil {
+		t.Fatalf("Prewarm: %v", err)
+	}
+	waitPrewarmDone(t, session)
+
+	tr := recorder.Finish()
+	if got := tr.Counter(trace.CounterPrewarmAttempts); got != 1 {
+		t.Fatalf("prewarm_attempts = %d, want 1", got)
+	}
+	var sawSpan bool
+	for _, span := range tr.Spans {
+		if span.Name == trace.SpanProviderPrewarm {
+			sawSpan = true
+		}
+	}
+	if !sawSpan {
+		t.Fatalf("missing %s span in %+v", trace.SpanProviderPrewarm, tr.Spans)
+	}
+}
+
+func TestTurnSessionStreamDelegatesToProvider(t *testing.T) {
+	handler := func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "text/event-stream")
+		_, _ = w.Write([]byte("data: {\"choices\":[{\"delta\":{\"content\":\"hello\"}}]}\n\n"))
+		_, _ = w.Write([]byte("data: [DONE]\n\n"))
+	}
+
+	direct := newTestProvider(t, handler)
+	viaSession := newTestProvider(t, handler)
+	session := openOptimizedSession(t, viaSession)
+
+	request := zeroruntime.CompletionRequest{
+		Messages: []zeroruntime.Message{{Role: zeroruntime.MessageRoleUser, Content: "hi"}},
+	}
+	collect := func(stream <-chan zeroruntime.StreamEvent) []zeroruntime.StreamEvent {
+		var events []zeroruntime.StreamEvent
+		for event := range stream {
+			events = append(events, event)
+		}
+		return events
+	}
+
+	directStream, err := direct.StreamCompletion(context.Background(), request)
+	if err != nil {
+		t.Fatalf("direct StreamCompletion: %v", err)
+	}
+	sessionStream, err := session.Stream(context.Background(), request)
+	if err != nil {
+		t.Fatalf("session Stream: %v", err)
+	}
+
+	directEvents := collect(directStream)
+	sessionEvents := collect(sessionStream)
+	if len(directEvents) != len(sessionEvents) {
+		t.Fatalf("event counts diverged: direct=%d session=%d", len(directEvents), len(sessionEvents))
+	}
+	for i := range directEvents {
+		if directEvents[i].Type != sessionEvents[i].Type || directEvents[i].Content != sessionEvents[i].Content {
+			t.Fatalf("event %d diverged: direct=%+v session=%+v", i, directEvents[i], sessionEvents[i])
+		}
+	}
+}
+
+func TestTurnSessionFingerprintIgnoresMessages(t *testing.T) {
+	provider := newTestProvider(t, func(http.ResponseWriter, *http.Request) {})
+	session := openOptimizedSession(t, provider)
+
+	tools := []zeroruntime.ToolDefinition{{Name: "read_file", Parameters: map[string]any{"type": "object"}}}
+	base := session.computeFingerprint(zeroruntime.CompletionRequest{
+		Messages:       []zeroruntime.Message{{Role: zeroruntime.MessageRoleUser, Content: "turn one"}},
+		Tools:          tools,
+		PromptCacheKey: "session-a",
+	})
+	grown := session.computeFingerprint(zeroruntime.CompletionRequest{
+		Messages: []zeroruntime.Message{
+			{Role: zeroruntime.MessageRoleUser, Content: "turn one"},
+			{Role: zeroruntime.MessageRoleAssistant, Content: "reply"},
+			{Role: zeroruntime.MessageRoleUser, Content: "turn two"},
+		},
+		Tools:          tools,
+		PromptCacheKey: "session-a",
+	})
+	if base != grown {
+		t.Fatal("fingerprint changed when only messages grew — conversation content must not be an input")
+	}
+}
+
+func TestTurnSessionFingerprintDriftOnCacheKey(t *testing.T) {
+	provider := newTestProvider(t, func(http.ResponseWriter, *http.Request) {})
+	session := openOptimizedSession(t, provider)
+
+	tools := []zeroruntime.ToolDefinition{{Name: "read_file", Parameters: map[string]any{"type": "object"}}}
+	base := session.computeFingerprint(zeroruntime.CompletionRequest{Tools: tools, PromptCacheKey: "session-a"})
+	changed := session.computeFingerprint(zeroruntime.CompletionRequest{Tools: tools, PromptCacheKey: "session-b"})
+	if base == changed {
+		t.Fatal("fingerprint did not drift on a changed prompt-cache key — it is a serialized request field")
+	}
+}
+
+func TestTurnSessionFingerprintDriftOnToolsAndEffort(t *testing.T) {
+	provider := newTestProvider(t, func(http.ResponseWriter, *http.Request) {})
+	session := openOptimizedSession(t, provider)
+
+	base := session.computeFingerprint(zeroruntime.CompletionRequest{
+		Tools: []zeroruntime.ToolDefinition{{Name: "read_file", Parameters: map[string]any{"type": "object"}}},
+	})
+	changedTools := session.computeFingerprint(zeroruntime.CompletionRequest{
+		Tools: []zeroruntime.ToolDefinition{{Name: "write_file", Parameters: map[string]any{"type": "object"}}},
+	})
+	if base == changedTools {
+		t.Fatal("fingerprint did not drift on a changed tool set")
+	}
+	changedEffort := session.computeFingerprint(zeroruntime.CompletionRequest{
+		Tools:           []zeroruntime.ToolDefinition{{Name: "read_file", Parameters: map[string]any{"type": "object"}}},
+		ReasoningEffort: "high",
+	})
+	if base == changedEffort {
+		t.Fatal("fingerprint did not drift on a changed reasoning effort")
+	}
+	// Unrecognized efforts normalize to omitted — same as empty.
+	droppedEffort := session.computeFingerprint(zeroruntime.CompletionRequest{
+		Tools:           []zeroruntime.ToolDefinition{{Name: "read_file", Parameters: map[string]any{"type": "object"}}},
+		ReasoningEffort: "xhigh",
+	})
+	if base != droppedEffort {
+		t.Fatal("an effort the wire would omit must fingerprint like no effort")
+	}
+}
+
+func TestTurnSessionFingerprintDriftOnToolDescription(t *testing.T) {
+	provider := newTestProvider(t, func(http.ResponseWriter, *http.Request) {})
+	session := openOptimizedSession(t, provider)
+
+	base := session.computeFingerprint(zeroruntime.CompletionRequest{
+		Tools: []zeroruntime.ToolDefinition{{Name: "read_file", Description: "Reads a file.", Parameters: map[string]any{"type": "object"}}},
+	})
+	changedDescription := session.computeFingerprint(zeroruntime.CompletionRequest{
+		Tools: []zeroruntime.ToolDefinition{{Name: "read_file", Description: "Reads a file, now with ranges.", Parameters: map[string]any{"type": "object"}}},
+	})
+	if base == changedDescription {
+		t.Fatal("fingerprint did not drift on a description-only tool change — descriptions are model-visible request bytes")
+	}
+}
+
+func TestTurnSessionFingerprintDriftOnToolReorder(t *testing.T) {
+	provider := newTestProvider(t, func(http.ResponseWriter, *http.Request) {})
+	session := openOptimizedSession(t, provider)
+
+	forward := session.computeFingerprint(zeroruntime.CompletionRequest{Tools: []zeroruntime.ToolDefinition{
+		{Name: "alpha", Parameters: map[string]any{"type": "object"}},
+		{Name: "beta", Parameters: map[string]any{"type": "object"}},
+	}})
+	reversed := session.computeFingerprint(zeroruntime.CompletionRequest{Tools: []zeroruntime.ToolDefinition{
+		{Name: "beta", Parameters: map[string]any{"type": "object"}},
+		{Name: "alpha", Parameters: map[string]any{"type": "object"}},
+	}})
+	if forward == reversed {
+		t.Fatal("fingerprint did not drift on a tool reorder — request serialization preserves tool order, so a reorder changes the wire bytes")
+	}
+}
+
+// TestTurnSessionPrewarmDoesNotStampProviderConnect guards the A/B metric: the
+// prewarm probe must appear ONLY under provider_prewarm — a provider_connect
+// stamp from the probe would contaminate the exact span the benchmark compares.
+func TestTurnSessionPrewarmDoesNotStampProviderConnect(t *testing.T) {
+	provider := newPrewarmTestProvider(t, func(w http.ResponseWriter, _ *http.Request) {
+		w.WriteHeader(http.StatusMethodNotAllowed)
+	})
+	recorder := trace.NewRecorder("session-test", "run-1", "test")
+	recorder.Start()
+	ctx := trace.WithContext(context.Background(), recorder)
+
+	session := openOptimizedSession(t, provider)
+	if err := session.Prewarm(ctx); err != nil {
+		t.Fatalf("Prewarm: %v", err)
+	}
+	waitPrewarmDone(t, session)
+
+	tr := recorder.Finish()
+	for _, span := range tr.Spans {
+		if span.Name == trace.SpanProviderConnect {
+			t.Fatalf("prewarm stamped a %s span — it must stay solely under %s", trace.SpanProviderConnect, trace.SpanProviderPrewarm)
+		}
+	}
+}
+
+// TestTurnSessionPrewarmSkippedWhenKeepAlivesDisabled verifies the probe is not
+// spent when the transport cannot retain idle connections (the macOS shared
+// transport): no request, no trace stamps, and the done channel still closes.
+func TestTurnSessionPrewarmSkippedWhenKeepAlivesDisabled(t *testing.T) {
+	var requests atomic.Int64
+	server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, _ *http.Request) {
+		requests.Add(1)
+		w.WriteHeader(http.StatusMethodNotAllowed)
+	}))
+	defer server.Close()
+
+	transport := http.DefaultTransport.(*http.Transport).Clone()
+	transport.DisableKeepAlives = true
+	provider, err := New(Options{
+		APIKey:     "sk-test",
+		BaseURL:    server.URL,
+		Model:      "gpt-test",
+		HTTPClient: &http.Client{Transport: transport},
+	})
+	if err != nil {
+		t.Fatalf("New: %v", err)
+	}
+
+	recorder := trace.NewRecorder("session-test", "run-1", "test")
+	recorder.Start()
+	ctx := trace.WithContext(context.Background(), recorder)
+
+	session := openOptimizedSession(t, provider)
+	if err := session.Prewarm(ctx); err != nil {
+		t.Fatalf("Prewarm: %v", err)
+	}
+	waitPrewarmDone(t, session)
+
+	if got := requests.Load(); got != 0 {
+		t.Fatalf("keep-alive-disabled transport still received %d probe requests, want 0", got)
+	}
+	tr := recorder.Finish()
+	if got := tr.Counter(trace.CounterPrewarmAttempts); got != 0 {
+		t.Fatalf("prewarm_attempts = %d, want 0 for a skipped probe", got)
+	}
+}
+
+func TestTurnSessionPrefixCounters(t *testing.T) {
+	handler := func(w http.ResponseWriter, _ *http.Request) {
+		w.Header().Set("Content-Type", "text/event-stream")
+		_, _ = w.Write([]byte("data: [DONE]\n\n"))
+	}
+	provider := newTestProvider(t, handler)
+	session := openOptimizedSession(t, provider)
+
+	recorder := trace.NewRecorder("session-test", "run-1", "test")
+	recorder.Start()
+	ctx := trace.WithContext(context.Background(), recorder)
+
+	stable := zeroruntime.CompletionRequest{Tools: []zeroruntime.ToolDefinition{{Name: "read_file", Parameters: map[string]any{"type": "object"}}}}
+	drain := func(request zeroruntime.CompletionRequest) {
+		stream, err := session.Stream(ctx, request)
+		if err != nil {
+			t.Fatalf("Stream: %v", err)
+		}
+		for range stream {
+		}
+	}
+
+	drain(stable)                                                                                                                                 // seeds — neither stable nor drift
+	drain(stable)                                                                                                                                 // stable
+	drain(zeroruntime.CompletionRequest{Tools: []zeroruntime.ToolDefinition{{Name: "write_file", Parameters: map[string]any{"type": "object"}}}}) // drift
+
+	tr := recorder.Finish()
+	if got := tr.Counter(trace.CounterPrefixStable); got != 1 {
+		t.Fatalf("prefix_stable = %d, want 1", got)
+	}
+	if got := tr.Counter(trace.CounterPrefixDrift); got != 1 {
+		t.Fatalf("prefix_drift = %d, want 1", got)
+	}
+}
+
+func TestTurnSessionCloseIdempotentAndCompactUnsupported(t *testing.T) {
+	provider := newTestProvider(t, func(http.ResponseWriter, *http.Request) {})
+	session := openOptimizedSession(t, provider)
+	if err := session.Close(); err != nil {
+		t.Fatalf("Close: %v", err)
+	}
+	if err := session.Close(); err != nil {
+		t.Fatalf("second Close: %v", err)
+	}
+	if _, err := session.Compact(context.Background(), zeroruntime.CompletionRequest{}); !errors.Is(err, zeroruntime.ErrCompactionUnsupported) {
+		t.Fatalf("Compact error = %v, want ErrCompactionUnsupported", err)
+	}
+}

internal/providers/turn_session.go

@@ -0,0 +1,72 @@
+package providers
+
+import (
+	"os"
+	"strings"
+
+	"github.com/Gitlawb/zero/internal/config"
+	"github.com/Gitlawb/zero/internal/providers/openai"
+	"github.com/Gitlawb/zero/internal/zeroruntime"
+)
+
+// openaiTurnSessionEnv gates the optimized OpenAI turn session (prewarm +
+// prefix telemetry). Default OFF: unset, "0", or "false" (any case) leave the
+// agent loop's default adapter path untouched. Set to any other non-empty
+// value (e.g. "1") to enable — the same boolean idiom as ZERO_FORMAT_ON_WRITE.
+const openaiTurnSessionEnv = "ZERO_OPENAI_TURN_SESSION"
+
+func openaiTurnSessionEnabled() bool {
+	value := strings.TrimSpace(os.Getenv(openaiTurnSessionEnv))
+	return value != "" && value != "0" && !strings.EqualFold(value, "false")
+}
+
+// OptimizedTurnSessions returns the gated optimized TurnSessionProvider for an
+// already-built provider, or (nil, false) when the gate is off or the profile
+// is ineligible. Callers leave agent Options.TurnSessionProvider nil on false,
+// so the loop's default wrap keeps today's behavior byte-identical.
+//
+// Eligibility is deliberately narrow: the resolved provider kind must be
+// official OpenAI (NOT openai-compatible gateways — the constructor merges the
+// two, so this branches on the resolved kind explicitly), not the ChatGPT
+// Codex catalog, and the provider value must be the concrete *openai.Provider
+// (a fake, a Codex provider, or nil falls back to the default path safely).
+// No base-URL check on top of the kind: prewarm and fingerprint telemetry are
+// harmless against any host, and kind==openai mirrors the existing
+// official-OpenAI precedent used for prompt_cache_key.
+func OptimizedTurnSessions(profile config.ProviderProfile, provider zeroruntime.Provider, options Options) (zeroruntime.TurnSessionProvider, bool) {
+	if !openaiTurnSessionEnabled() {
+		return nil, false
+	}
+	resolved, err := resolveProfile(profile, options)
+	if err != nil || resolved.providerKind != config.ProviderKindOpenAI || isCodexCatalog(profile, resolved) {
+		return nil, false
+	}
+	concrete, ok := provider.(*openai.Provider)
+	if !ok {
+		return nil, false
+	}
+	caps, err := resolveCapabilities(profile, options)
+	if err != nil {
+		// resolveProfile above already succeeded, so this is effectively
+		// unreachable — but never ship an optimized session with unknown
+		// capabilities; the default path is the safe answer.
+		return nil, false
+	}
+	return openai.NewTurnSessionProvider(concrete, caps), true
+}
+
+// DefaultTurnSessions wraps an already-built provider in the DEFAULT (no-op)
+// turn-session adapter with the profile's resolved capability projection. It is
+// the non-optimized sibling of OptimizedTurnSessions for callers that still
+// want capabilities populated — e.g. the mid-run model-switch fallback when the
+// switched model is not eligible for the optimized session. Capability
+// resolution failure degrades to an unknown (zero) projection rather than
+// failing the wrap: the default session has no behavior that depends on
+// capabilities, so a swap must never be blocked by a projection error.
+func DefaultTurnSessions(profile config.ProviderProfile, provider zeroruntime.Provider, options Options) zeroruntime.TurnSessionProvider {
+	caps, err := resolveCapabilities(profile, options)
+	if err != nil {
+		caps = zeroruntime.ProviderCapabilities{}
+	}
+	return zeroruntime.NewProviderTurnSessionProvider(provider, caps)
+}

internal/providers/turn_session_gate_test.go

@@ -0,0 +1,128 @@
+package providers
+
+import (
+	"context"
+	"testing"
+
+	"github.com/Gitlawb/zero/internal/config"
+	"github.com/Gitlawb/zero/internal/zeroruntime"
+)
+
+func openaiEligibleProfile() config.ProviderProfile {
+	return config.ProviderProfile{
+		Name:         "official-openai",
+		ProviderKind: config.ProviderKindOpenAI,
+		BaseURL:      "https://provider.example/v1",
+		APIKey:       "sk-gate-test",
+		Model:        "pr8-unregistered-model",
+	}
+}
+
+func buildGateProvider(t *testing.T, profile config.ProviderProfile) zeroruntime.Provider {
+	t.Helper()
+	provider, err := New(profile, Options{UserAgent: "zero-gate-test"})
+	if err != nil {
+		t.Fatalf("New: %v", err)
+	}
+	return provider
+}
+
+type fakeGateProvider struct{}
+
+func (fakeGateProvider) StreamCompletion(context.Context, zeroruntime.CompletionRequest) (<-chan zeroruntime.StreamEvent, error) {
+	return nil, nil
+}
+
+func TestOptimizedTurnSessionsDefaultOff(t *testing.T) {
+	// No env set: the gate must be off even for a fully eligible profile.
+	profile := openaiEligibleProfile()
+	tsp, ok := OptimizedTurnSessions(profile, buildGateProvider(t, profile), Options{})
+	if ok || tsp != nil {
+		t.Fatal("optimized turn sessions must be disabled by default")
+	}
+}
+
+func TestOptimizedTurnSessionsFalseyValues(t *testing.T) {
+	profile := openaiEligibleProfile()
+	provider := buildGateProvider(t, profile)
+	for _, value := range []string{"0", "false", "FALSE", "False", " ", ""} {
+		t.Setenv(openaiTurnSessionEnv, value)
+		if _, ok := OptimizedTurnSessions(profile, provider, Options{}); ok {
+			t.Fatalf("gate enabled for falsey value %q", value)
+		}
+	}
+	for _, value := range []string{"1", "true", "on"} {
+		t.Setenv(openaiTurnSessionEnv, value)
+		if _, ok := OptimizedTurnSessions(profile, provider, Options{}); !ok {
+			t.Fatalf("gate disabled for truthy value %q", value)
+		}
+	}
+}
+
+func TestOptimizedTurnSessionsOpenAIEligible(t *testing.T) {
+	t.Setenv(openaiTurnSessionEnv, "1")
+	profile := openaiEligibleProfile()
+	tsp, ok := OptimizedTurnSessions(profile, buildGateProvider(t, profile), Options{})
+	if !ok || tsp == nil {
+		t.Fatal("expected the optimized session for an official-OpenAI profile with the gate on")
+	}
+	session, err := tsp.OpenTurnSession(context.Background())
+	if err != nil {
+		t.Fatalf("OpenTurnSession: %v", err)
+	}
+	if err := session.Close(); err != nil {
+		t.Fatalf("Close: %v", err)
+	}
+	if caps := tsp.Capabilities(); caps.Model != "pr8-unregistered-model" {
+		t.Fatalf("Capabilities().Model = %q, want the resolved api model", caps.Model)
+	}
+}
+
+func TestOptimizedTurnSessionsRejectsCompatible(t *testing.T) {
+	t.Setenv(openaiTurnSessionEnv, "1")
+	profile := openaiEligibleProfile()
+	profile.ProviderKind = config.ProviderKindOpenAICompatible
+	if _, ok := OptimizedTurnSessions(profile, buildGateProvider(t, profile), Options{}); ok {
+		t.Fatal("openai-compatible gateways must not get the optimized session")
+	}
+}
+
+func TestOptimizedTurnSessionsRejectsCodexCatalog(t *testing.T) {
+	t.Setenv(openaiTurnSessionEnv, "1")
+	profile := openaiEligibleProfile()
+	profile.CatalogID = "chatgpt"
+	// Build the provider from the base profile: the point under test is the
+	// catalog check, not Codex construction (which needs OAuth wiring).
+	if _, ok := OptimizedTurnSessions(profile, buildGateProvider(t, openaiEligibleProfile()), Options{}); ok {
+		t.Fatal("the ChatGPT Codex catalog must not get the optimized session")
+	}
+}
+
+func TestDefaultTurnSessionsPreservesResolvedCapabilities(t *testing.T) {
+	profile := openaiEligibleProfile()
+	tsp := DefaultTurnSessions(profile, buildGateProvider(t, profile), Options{})
+	if tsp == nil {
+		t.Fatal("DefaultTurnSessions returned nil")
+	}
+	caps := tsp.Capabilities()
+	if caps.Model != "pr8-unregistered-model" {
+		t.Fatalf("Capabilities().Model = %q, want the resolved api model", caps.Model)
+	}
+	session, err := tsp.OpenTurnSession(context.Background())
+	if err != nil {
+		t.Fatalf("OpenTurnSession: %v", err)
+	}
+	if err := session.Close(); err != nil {
+		t.Fatalf("Close: %v", err)
+	}
+}
+
+func TestOptimizedTurnSessionsRejectsForeignProviderValue(t *testing.T) {
+	t.Setenv(openaiTurnSessionEnv, "1")
+	if _, ok := OptimizedTurnSessions(openaiEligibleProfile(), fakeGateProvider{}, Options{}); ok {
+		t.Fatal("a non-*openai.Provider value must fall back to the default path")
+	}
+	if _, ok := OptimizedTurnSessions(openaiEligibleProfile(), nil, Options{}); ok {
+		t.Fatal("a nil provider must fall back to the default path")
+	}
+}

internal/sandbox/analyzer.go

@@ -65,6 +65,76 @@ func AnalyzeCommand(script string) AnalysisResult {
 	return result
 }
 
+// astCommandFields parses command with the shell parser and returns each simple
+// command as its literal field slice (program + args as text), resolving the
+// real command positions across quoting, command substitution, subshells, and
+// newline separators — the constructs the hand-written splitter in
+// safe_command.go mis-handles (issue #473). It returns nil when the command
+// cannot be parsed (e.g. a Windows cmd.exe string), so callers fall through to
+// the regex path rather than hard-blocking.
+func astCommandFields(command string) [][]string {
+	file, err := syntax.NewParser().Parse(strings.NewReader(command), "")
+	if err != nil {
+		return nil
+	}
+	var commands [][]string
+	syntax.Walk(file, func(node syntax.Node) bool {
+		call, ok := node.(*syntax.CallExpr)
+		if !ok || len(call.Args) == 0 {
+			return true
+		}
+		// EVERY word must be a static literal, not just the program name.
+		// wordText keeps only the literal/quoted parts of a word and silently
+		// drops expansions, so a dynamic word anywhere in the call reconstructs
+		// to something the shell will never run: `$(printf foo)vim` (runs as
+		// `foovim`) collapses to "vim", and `git $(printf foo)rebase -i` (runs
+		// as `git foorebase -i`, non-interactive) collapses to `git rebase -i`
+		// and fabricates an interactive match. Since the runtime value of an
+		// expansion is unknowable here, skip the whole call rather than classify
+		// a lossy reconstruction. Skipping is the safe direction: the
+		// hand-written passes above already ran, and a missed detection falls
+		// through to the normal permission prompt instead of hard-blocking a
+		// command the user never wrote.
+		for _, word := range call.Args {
+			if !isLiteralWord(word) {
+				return true
+			}
+		}
+		fields := make([]string, 0, len(call.Args))
+		for _, word := range call.Args {
+			fields = append(fields, wordText(word))
+		}
+		commands = append(commands, fields)
+		return true
+	})
+	return commands
+}
+
+// isLiteralWord reports whether every part of word is a static literal (bare or
+// quoted). A word containing a command substitution, parameter/arithmetic
+// expansion, process substitution, etc. is dynamic — its runtime value is
+// unknown, so its wordText (a partial literal) must not be trusted as a program
+// name.
+func isLiteralWord(word *syntax.Word) bool {
+	if word == nil {
+		return false
+	}
+	for _, part := range word.Parts {
+		switch typed := part.(type) {
+		case *syntax.Lit, *syntax.SglQuoted:
+		case *syntax.DblQuoted:
+			for _, inner := range typed.Parts {
+				if _, ok := inner.(*syntax.Lit); !ok {
+					return false
+				}
+			}
+		default:
+			return false
+		}
+	}
+	return true
+}
+
 // analyzeInto parses script and folds its interactive/destructive/network usage
 // into result, sharing seen so program names are de-duplicated across recursion.
 func analyzeInto(script string, result *AnalysisResult, seen map[string]bool, depth int) {

internal/sandbox/auto_allow_bash_test.go

@@ -93,6 +93,40 @@ func TestSandboxedBashAutoAllowDoesNotAffectNonShell(t *testing.T) {
 	}
 }
 
+// Forged re-entrancy markers (ambient ZERO_SANDBOXED=1 + ZERO_SANDBOX_BACKEND at
+// an unsandboxed process boundary) must NOT let a shell command combine an
+// auto-allow decision with an unwrapped pass-through plan (issue #727). The
+// runner skips wrapping when the markers are present, so the policy must stop
+// auto-allowing on the basis of an "active sandbox" — the command takes the
+// normal prompt instead. Without the markers the identical setup auto-allows
+// (TestSandboxedBashAutoAllowedWhenSandboxActive), so this is the before/after.
+func TestForgedReentrancyMarkersDoNotAutoAllowUnwrappedShell(t *testing.T) {
+	t.Setenv(EnvSandboxed, "1")
+	t.Setenv(EnvSandboxBackend, string(BackendLinuxBwrap))
+	root := t.TempDir()
+	engine := NewEngine(EngineOptions{
+		WorkspaceRoot: root,
+		Policy:        DefaultPolicy(),
+		Backend:       nativeWrappingBackend,
+	})
+
+	// Policy side: the shell command must NOT be auto-allowed.
+	decision := engine.Evaluate(context.Background(), bashRequest())
+	if decision.Action != ActionPrompt || decision.AutoAllowed {
+		t.Fatalf("forged markers must not auto-allow the shell command, got %#v", decision)
+	}
+
+	// Planning side: the same command is an unwrapped pass-through. Together with
+	// the prompt above, this proves auto-allow and unwrapped can no longer combine.
+	plan, err := engine.BuildCommandPlan(CommandSpec{Name: "/bin/sh", Args: []string{"-c", "echo hi"}, Dir: root})
+	if err != nil {
+		t.Fatalf("BuildCommandPlan: %v", err)
+	}
+	if plan.Wrapped {
+		t.Fatalf("re-entrancy plan must be an unwrapped pass-through, got wrapped=%v", plan.Wrapped)
+	}
+}
+
 // TestShellSandboxActive reports correctly across backends and policy modes.
 func TestShellSandboxActive(t *testing.T) {
 	root := t.TempDir()

internal/sandbox/engine.go

@@ -232,6 +232,15 @@ func (engine *Engine) shellSandboxActive(policy Policy) bool {
 	if policy.Mode == ModeDisabled {
 		return false
 	}
+	// Re-entrancy skips wrapping: when the process carries the sandbox markers,
+	// the runner returns a pass-through (unwrapped) plan, so the native sandbox
+	// is NOT the boundary for THIS command. Because the markers are ambient and
+	// forgeable at an unsandboxed process boundary (issue #727), auto-allowing a
+	// shell command on that basis would run it unwrapped with no prompt. Mirror
+	// the runner: report inactive so the command takes the normal approval path.
+	if IsAlreadySandboxed() {
+		return false
+	}
 	backend := engine.backend
 	if !(backend.Available && backend.Executable != "" && backend.CommandWrapping && backend.NativeIsolation) {
 		return false

internal/sandbox/safe_command.go

@@ -219,9 +219,65 @@ func DetectInteractiveCommand(command string, goos string) InteractiveCommandRes
 		}
 	}
 
+	// AST second opinion (issue #473): the hand-written segment split above
+	// mis-handles interactive programs hidden by unusual quoting, command
+	// substitution, subshells, or newline separators. Re-extract the real simple
+	// commands from the parsed shell tree and apply the SAME full pipeline
+	// (interactive segments, sh -c payload recursion, per-program lookup), so a
+	// bypass is caught while every program stays classified exactly as before
+	// (ssh with a trailing command, python -c, etc. remain allowed). A command
+	// the parser cannot handle (Windows cmd.exe, obfuscation) yields no commands
+	// and falls through unchanged — the guard never hard-blocks on a parse error.
+	for _, fields := range astCommandFields(command) {
+		if result, ok := inspectCommandFields(fields, goos); ok {
+			return result
+		}
+	}
+
 	return InteractiveCommandResult{}
 }
 
+// inspectCommandFields applies the full interactive-detection pipeline to one
+// already-split command's fields: the multi-word interactive segments, then the
+// `sh -c <payload>` recursion, then the per-program lookup with its
+// non-interactive suppressions (hasNonInteractiveFlag covers REPL flags and
+// trailing-command clients like ssh). It mirrors the hand-written passes above
+// so an AST-extracted command is classified identically to a plainly-split one.
+func inspectCommandFields(fields []string, goos string) (InteractiveCommandResult, bool) {
+	body := strings.ToLower(commandBody(fields))
+	for _, seg := range interactiveSegments {
+		if body == seg.match || strings.HasPrefix(body, seg.match+" ") {
+			suggestion := seg.suggestion
+			if goos == "windows" && seg.windowsSuggestion != "" {
+				suggestion = seg.windowsSuggestion
+			}
+			return InteractiveCommandResult{Interactive: true, Command: seg.command, Reason: seg.reason, Suggestion: suggestion}, true
+		}
+	}
+	first := firstProgram(fields)
+	if first == "" {
+		return InteractiveCommandResult{}, false
+	}
+	if payload := shellDashCPayload(first, fields); payload != "" {
+		if inner := DetectInteractiveCommand(payload, goos); inner.Interactive {
+			return inner, true
+		}
+		return InteractiveCommandResult{}, false
+	}
+	program, ok := interactivePrograms[first]
+	if !ok || (program.windowsOnly && goos != "windows") {
+		return InteractiveCommandResult{}, false
+	}
+	if hasNonInteractiveFlag(first, fields) {
+		return InteractiveCommandResult{}, false
+	}
+	suggestion := program.suggestion
+	if goos == "windows" && program.windowsSuggestion != "" {
+		suggestion = program.windowsSuggestion
+	}
+	return InteractiveCommandResult{Interactive: true, Command: first, Reason: program.reason, Suggestion: suggestion}, true
+}
+
 // wrapperPrograms are launcher prefixes that precede the real program. After
 // one of these we keep scanning for the actual executable.
 var wrapperPrograms = map[string]bool{

internal/sandbox/safe_command_test.go

@@ -406,3 +406,87 @@ func TestDetectInteractiveMongoEvalAndFullPaths(t *testing.T) {
 		}
 	}
 }
+
+// The hand-written segment splitter misses interactive programs hidden by
+// constructs only a real shell parser resolves (a newline separator collapsed
+// to a space; a brace group that shifts the real command position). The AST
+// second opinion must catch them (issue #473).
+func TestDetectInteractiveCommandCatchesParserBypasses(t *testing.T) {
+	for _, cmd := range []string{
+		"echo hi\nvim file.txt", // newline separator collapsed to a space by the regex path
+		"{ vim file.txt; }",     // brace group hides the real program position
+	} {
+		got := DetectInteractiveCommand(cmd, "linux")
+		if !got.Interactive {
+			t.Errorf("DetectInteractiveCommand(%q) = not interactive, want caught (parser bypass)", cmd)
+			continue
+		}
+		if got.Command != "vim" {
+			t.Errorf("DetectInteractiveCommand(%q).Command = %q, want vim", cmd, got.Command)
+		}
+	}
+}
+
+// The AST second opinion must not flag an interactive program NAME that appears
+// only inside a quoted argument (not a real command position) — the parser
+// distinguishes program from argument, so this must stay non-interactive.
+func TestDetectInteractiveCommandNoFalsePositiveOnQuotedArgument(t *testing.T) {
+	got := DetectInteractiveCommand(`echo "please run vim later"`, "linux")
+	if got.Interactive {
+		t.Fatalf("interactive name inside a quoted argument must not be flagged, got %#v", got)
+	}
+}
+
+// The AST second opinion must run the FULL pipeline — multi-word interactive
+// segments and sh -c payload recursion, not just the per-program lookup — so a
+// grouped/nested bypass is caught; and it must not fabricate a program name from
+// a command substitution concatenated with a literal (issue #473, review).
+func TestDetectInteractiveCommandASTPipelineBoundaries(t *testing.T) {
+	interactive := []struct{ cmd, wantCmd string }{
+		{"{ git rebase -i HEAD~1; }", "git rebase -i"}, // multi-word segment in a brace group
+		{"{ sh -c 'vim file.txt'; }", "vim"},           // sh -c payload in a brace group
+	}
+	for _, tc := range interactive {
+		got := DetectInteractiveCommand(tc.cmd, "linux")
+		if !got.Interactive || got.Command != tc.wantCmd {
+			t.Errorf("DetectInteractiveCommand(%q) = %+v, want interactive Command=%q", tc.cmd, got, tc.wantCmd)
+		}
+	}
+}
+
+// The AST pass must not fabricate a program name from a dynamic (non-literal)
+// program word: `$(printf '%s' foo)vim` runs as `foovim`, not vim. Tested at the
+// extractor so the hand-written pass (which pre-empts DetectInteractiveCommand
+// with its own, separate handling of that construct) does not mask the guard.
+func TestAstCommandFieldsSkipsDynamicProgram(t *testing.T) {
+	for _, fields := range astCommandFields("$(printf '%s' foo)vim file.txt") {
+		if firstProgram(fields) == "vim" {
+			t.Fatalf("astCommandFields fabricated a vim command from a substitution: %v", fields)
+		}
+	}
+}
+
+// The literalness guard covers every word, not just the program name. A dynamic
+// ARGUMENT is dropped by wordText the same way a dynamic program word is, so
+// `git $(printf foo)rebase -i HEAD~1` (which runs as `git foorebase -i`, a
+// non-existent and non-interactive subcommand) would otherwise be reconstructed
+// as `git rebase -i` and blocked — a false positive on a command the user never
+// wrote. Asserted through DetectInteractiveCommand as well as the extractor,
+// because the whole point is what the caller ends up blocking.
+func TestAstCommandFieldsSkipsDynamicArgument(t *testing.T) {
+	const dynamic = "{ git $(printf foo)rebase -i HEAD~1 ; }"
+	for _, fields := range astCommandFields(dynamic) {
+		if firstProgram(fields) == "git" {
+			t.Fatalf("astCommandFields reconstructed a git command from a dynamic argument: %v", fields)
+		}
+	}
+	if got := DetectInteractiveCommand(dynamic, "linux"); got.Interactive {
+		t.Errorf("DetectInteractiveCommand(%q) = %+v, want not interactive", dynamic, got)
+	}
+	// The genuine form must still be detected — the guard skips lossy
+	// reconstructions, it does not weaken real detection.
+	const genuine = "{ git rebase -i HEAD~1 ; }"
+	if got := DetectInteractiveCommand(genuine, "linux"); !got.Interactive || got.Command != "git rebase -i" {
+		t.Errorf("DetectInteractiveCommand(%q) = %+v, want interactive Command=%q", genuine, got, "git rebase -i")
+	}
+}

internal/sessions/exec_session.go

@@ -89,7 +89,7 @@ func PrepareExec(options PrepareExecOptions) (PreparedExec, error) {
 	if resumeID != "" || options.ResumeLatest {
 		sessionID := resumeID
 		if sessionID == "" && options.ResumeLatest {
-			latest, err := store.Latest()
+			latest, err := store.LatestResumable()
 			if err != nil {
 				return PreparedExec{}, err
 			}
@@ -105,6 +105,9 @@ func PrepareExec(options PrepareExecOptions) (PreparedExec, error) {
 		if session == nil {
 			return PreparedExec{}, ExecError{"Zero session not found: " + sessionID}
 		}
+		if !IsResumableKind(session.SessionKind) {
+			return PreparedExec{}, ExecError{"Zero session is not resumable: " + sessionID}
+		}
 		contextEvents, err := readExecContextEvents(store, session.SessionID)
 		if err != nil {
 			return PreparedExec{}, err

internal/sessions/store.go

@@ -52,6 +52,7 @@ type SessionKind string
 const (
 	SessionKindFork      SessionKind = "fork"
 	SessionKindChild     SessionKind = "child"
+	SessionKindSide      SessionKind = "side"
 	SessionKindSpecDraft SessionKind = "spec-draft"
 	SessionKindSpecImpl  SessionKind = "spec-impl"
 )
@@ -125,11 +126,13 @@ type CreateInput struct {
 }
 
 type ForkInput struct {
-	SessionID string
-	Title     string
-	Cwd       string
-	ModelID   string
-	Provider  string
+	SessionID   string
+	SessionKind SessionKind
+	Title       string
+	Cwd         string
+	ModelID     string
+	Provider    string
+	Tag         string
 }
 
 type ChildInput struct {
@@ -354,10 +357,9 @@ func (store *Store) Latest() (*Metadata, error) {
 }
 
 // IsResumableKind reports whether a session kind represents a standalone,
-// user-resumable conversation rather than an agent sub-run. Regular ("") and
-// user fork sessions are resumable; child (specialist/sub-agent) and spec
-// draft/impl sessions are not — each agent task and /spec run creates one, so
-// listing them in the resume picker floods it with non-conversation entries.
+// user-resumable conversation rather than a transient or agent-owned sub-run.
+// Regular ("") and user fork sessions are resumable; side, child, and spec
+// sessions are not, so they do not flood the resume picker.
 func IsResumableKind(kind SessionKind) bool {
 	switch kind {
 	case "", SessionKindFork:
@@ -418,13 +420,21 @@ func (store *Store) Fork(parentSessionID string, input ForkInput) (Metadata, err
 	if title == "" && parent.Title != "" {
 		title = parent.Title + " (fork)"
 	}
+	kind := input.SessionKind
+	if kind == "" {
+		kind = SessionKindFork
+	}
+	if kind != SessionKindFork && kind != SessionKindSide {
+		return Metadata{}, fmt.Errorf("invalid zero fork session kind %q", kind)
+	}
 	fork, err := store.Create(CreateInput{
 		SessionID:          input.SessionID,
-		SessionKind:        SessionKindFork,
+		SessionKind:        kind,
 		Title:              title,
 		Cwd:                firstNonEmpty(input.Cwd, parent.Cwd),
 		ModelID:            firstNonEmpty(input.ModelID, parent.ModelID),
 		Provider:           firstNonEmpty(input.Provider, parent.Provider),
+		Tag:                input.Tag,
 		ParentSessionID:    parent.SessionID,
 		RootSessionID:      firstNonEmpty(parent.RootSessionID, parent.SessionID),
 		ForkedFromEventID:  last.ID,

internal/sessions/store_test.go

@@ -181,6 +181,32 @@ func TestStoreForkCopiesEventsAndLineage(t *testing.T) {
 	}
 }
 
+func TestStoreForkSupportsNonResumableSideSession(t *testing.T) {
+	store := NewStore(StoreOptions{RootDir: t.TempDir()})
+	parent, err := store.Create(CreateInput{SessionID: "parent", Title: "Parent"})
+	if err != nil {
+		t.Fatalf("Create parent: %v", err)
+	}
+	if _, err := store.AppendEvent(parent.SessionID, AppendEventInput{Type: EventMessage, Payload: map[string]string{"content": "context"}}); err != nil {
+		t.Fatalf("AppendEvent: %v", err)
+	}
+
+	side, err := store.Fork(parent.SessionID, ForkInput{
+		SessionID:   "side",
+		SessionKind: SessionKindSide,
+		Tag:         "btw",
+	})
+	if err != nil {
+		t.Fatalf("Fork side: %v", err)
+	}
+	if side.SessionKind != SessionKindSide || side.Tag != "btw" || side.ParentSessionID != parent.SessionID {
+		t.Fatalf("side metadata = %#v", side)
+	}
+	if IsResumableKind(side.SessionKind) {
+		t.Fatal("side sessions must not appear as resumable conversations")
+	}
+}
+
 func TestStoreForkSkipsUsageEvents(t *testing.T) {
 	store := NewStore(StoreOptions{RootDir: t.TempDir(), Now: fixedClock("2026-06-04T11:00:00Z")})
 	parent, err := store.Create(CreateInput{SessionID: "parent", Title: "Parent"})
@@ -381,6 +407,7 @@ func TestListAndLatestResumableExcludeSubRuns(t *testing.T) {
 	mk(SessionKindSpecImpl, "spec-impl-1")
 	newestResumable := mk("", "conversation-2") // newest standalone conversation
 	mk(SessionKindChild, "child-2")             // newer overall, but a sub-run
+	mk(SessionKindSide, "side-1")               // newer overall, but non-resumable
 
 	resumable, err := store.ListResumable()
 	if err != nil {
@@ -404,7 +431,7 @@ func TestListAndLatestResumableExcludeSubRuns(t *testing.T) {
 		if latest != nil {
 			got = latest.SessionID
 		}
-		t.Fatalf("LatestResumable = %s, want newest resumable %s (must skip the newer child)", got, newestResumable.SessionID)
+		t.Fatalf("LatestResumable = %s, want newest resumable %s (must skip newer child and side sessions)", got, newestResumable.SessionID)
 	}
 }
 
@@ -535,6 +562,9 @@ func TestPrepareExecSessionResolvesResumeAndFork(t *testing.T) {
 	if _, err := store.AppendEvent("latest", AppendEventInput{Type: EventMessage, Payload: map[string]any{"content": "previous answer"}}); err != nil {
 		t.Fatal(err)
 	}
+	if _, err := store.Create(CreateInput{SessionID: "newer-side", SessionKind: SessionKindSide}); err != nil {
+		t.Fatal(err)
+	}
 
 	prepared, err := PrepareExec(PrepareExecOptions{Store: store, ResumeLatest: true})
 	if err != nil {
@@ -546,6 +576,9 @@ func TestPrepareExecSessionResolvesResumeAndFork(t *testing.T) {
 	if got := FormatExecPrompt("continue", prepared); got == "continue" || !strings.Contains(got, "previous answer") {
 		t.Fatalf("expected session context in prompt, got %q", got)
 	}
+	if _, err := PrepareExec(PrepareExecOptions{Store: store, Resume: "newer-side"}); err == nil || !strings.Contains(err.Error(), "Zero session is not resumable: newer-side") {
+		t.Fatalf("explicit side-session resume error = %v, want non-resumable rejection", err)
+	}
 
 	forked, err := PrepareExec(PrepareExecOptions{Store: store, Fork: "latest", SessionID: "forked"})
 	if err != nil {

internal/tools/exec_command.go

@@ -707,6 +707,18 @@ type writeStdinTool struct {
 	manager *execSessionManager
 }
 
+// UnknownExecSessionError is the result returned when write_stdin targets a
+// session_id with no live exec session. The stable recovery guidance leads and
+// the numeric id trails on purpose: the agent's repeated-failure guard keys on
+// a normalized, truncated prefix of the error string, so keeping the id out of
+// that prefix makes a model probing ids 1, 2, 3, … produce ONE signature that
+// finally trips the halt instead of a fresh signature per id — while also
+// telling the model how to actually recover. See TestUnknownExecSessionErrorSignatureIsIDInvariant
+// in internal/agent, which pins this invariant against the real guard.
+func UnknownExecSessionError(sessionID int) string {
+	return fmt.Sprintf("Error: write_stdin needs a session_id returned by a still-running exec_command; do not guess or probe session ids. Start a process with exec_command, or use write_file/edit_file/apply_patch for file changes. (no live session %d)", sessionID)
+}
+
 func (writeStdinTool) outputCategory(map[string]any) outputCategory {
 	return outputCategoryProcess
 }
@@ -722,7 +734,7 @@ func NewWriteStdinTool(manager *execSessionManager) Tool {
 			parameters: Schema{
 				Type: "object",
 				Properties: map[string]PropertySchema{
-					"session_id":        {Type: "integer", Description: "Identifier of the running unified exec session."},
+					"session_id":        {Type: "integer", Description: "Identifier of a running unified exec session, as returned by exec_command. Never guess or probe ids.", Minimum: intPtr(1)},
 					"chars":             {Type: "string", Description: "Bytes to write to stdin. Defaults to empty, which polls without writing.", Default: ""},
 					"yield_time_ms":     {Type: "integer", Description: "Wait before yielding output. Non-empty writes default to 250 ms and cap at 30000 ms; empty polls wait 5000-300000 ms by default.", Default: defaultPollYieldTimeMS, Minimum: intPtr(1), Maximum: intPtr(maxPollYieldTimeMS)},
 					"max_output_tokens": {Type: "integer", Description: "Output token budget. Defaults to 10000 tokens; larger requests may be capped by policy.", Default: defaultMaxOutputTokens, Minimum: intPtr(1), Maximum: intPtr(maxExecOutputTokenRequest)},
@@ -763,12 +775,20 @@ func (tool writeStdinTool) Run(ctx context.Context, args map[string]any) Result
 }
 
 func (tool writeStdinTool) RunWithOptions(ctx context.Context, args map[string]any, _ RunOptions) Result {
-	if value, ok := args["session_id"]; !ok || value == nil {
-		return errorResult("Error: Invalid arguments for write_stdin: session_id is required")
-	}
-	sessionID, err := intArg(args, "session_id", 0, 1, 0)
-	if err != nil {
-		return errorResult("Error: Invalid arguments for write_stdin: " + err.Error())
+	// A missing, non-integer, or < 1 session_id all mean the same thing: the model
+	// has no live session to write to. Route them to the SAME recovery guidance the
+	// no-live-session case uses (UnknownExecSessionError) instead of a terse
+	// "session_id must be at least 1". The terse error gives no way forward and, by
+	// naming the minimum, nudges the model to try 1, 2, 3... — the exact id-probing
+	// #749 works to suppress. The recovery message instead tells it to start a
+	// session or edit files directly, and because that message is id-invariant, the
+	// missing/zero/non-integer entry points and id-probing collapse to ONE
+	// repeated-failure signature, so any mix of them accumulates toward the halt
+	// rather than resetting the streak on each class of mistake.
+	value, present := args["session_id"]
+	sessionID, sessionErr := intArg(args, "session_id", 0, 1, 0)
+	if !present || value == nil || sessionErr != nil {
+		return errorResult(UnknownExecSessionError(sessionID))
 	}
 	chars, err := stringArgWithEmpty(args, "chars", "", false, true)
 	if err != nil {
@@ -784,7 +804,7 @@ func (tool writeStdinTool) RunWithOptions(ctx context.Context, args map[string]a
 	}
 	session, ok := tool.manager.get(sessionID)
 	if !ok {
-		return errorResult(fmt.Sprintf("Error: Unknown exec session_id %d.", sessionID))
+		return errorResult(UnknownExecSessionError(sessionID))
 	}
 	session.touch()
 	interrupted := false

internal/tools/exec_command_test.go

@@ -881,7 +881,7 @@ func TestRegistryHonorsWriteStdinArgumentPermission(t *testing.T) {
 	registry.Register(NewWriteStdinTool(newExecSessionManager()))
 
 	poll := registry.Run(context.Background(), WriteStdinToolName, map[string]any{"session_id": 9999})
-	if poll.Status != StatusError || !strings.Contains(poll.Output, "Unknown exec session_id") {
+	if poll.Status != StatusError || !strings.Contains(poll.Output, "still-running exec_command") {
 		t.Fatalf("empty poll should reach tool without permission prompt, got status=%s output=%q", poll.Status, poll.Output)
 	}
 
@@ -901,24 +901,57 @@ func TestWriteStdinReportsUnknownSession(t *testing.T) {
 	if result.Status != StatusError {
 		t.Fatalf("status = %s, want error", result.Status)
 	}
-	if !strings.Contains(result.Output, "Unknown exec session_id 1234") {
-		t.Fatalf("unexpected output: %q", result.Output)
+	// The message must lead with recovery guidance and still identify the id.
+	if !strings.Contains(result.Output, "still-running exec_command") {
+		t.Fatalf("unknown-session error must carry recovery guidance, got: %q", result.Output)
+	}
+	if !strings.Contains(result.Output, "1234") {
+		t.Fatalf("unknown-session error must name the offending id, got: %q", result.Output)
 	}
 }
 
+// The runtime rejects session_id < 1 (intArg min 1); the advertised schema must
+// say the same so a model sees the constraint before it probes id 0.
+func TestWriteStdinSchemaPinsSessionIDMinimum(t *testing.T) {
+	schema := NewWriteStdinTool(nil).(writeStdinTool).parameters
+	prop, ok := schema.Properties["session_id"]
+	if !ok {
+		t.Fatal("session_id property missing from write_stdin schema")
+	}
+	if prop.Minimum == nil || *prop.Minimum != 1 {
+		t.Fatalf("session_id schema Minimum = %v, want 1 to match the runtime floor", prop.Minimum)
+	}
+}
+
+// A missing, zero, negative, or non-integer session_id all mean the model has no
+// live session, so write_stdin returns the SAME recovery guidance the no-live-
+// session case uses (start a session with exec_command, or edit files directly),
+// not a terse "session_id must be at least 1" that gives no way forward and names
+// a minimum that nudges the model to probe ids 1, 2, 3... Sharing one id-invariant
+// message also keeps a single repeated-failure signature, so any mix of these
+// mistakes accumulates toward the halt instead of resetting the streak (#749).
 func TestWriteStdinRequiresPositiveSessionID(t *testing.T) {
 	tool := NewWriteStdinTool(newExecSessionManager())
-	for _, args := range []map[string]any{
-		{},
-		{"session_id": 0},
+	want := UnknownExecSessionError(0)
+	for name, args := range map[string]map[string]any{
+		"missing":     {},
+		"nil":         {"session_id": nil},
+		"zero":        {"session_id": 0},
+		"negative":    {"session_id": -3},
+		"non-integer": {"session_id": "abc"},
 	} {
-		result := tool.Run(context.Background(), args)
-		if result.Status != StatusError {
-			t.Fatalf("Run(%#v) status = %s, want error", args, result.Status)
-		}
-		if !strings.Contains(result.Output, "Invalid arguments for write_stdin") {
-			t.Fatalf("Run(%#v) output = %q, want invalid arguments", args, result.Output)
-		}
+		t.Run(name, func(t *testing.T) {
+			result := tool.Run(context.Background(), args)
+			if result.Status != StatusError {
+				t.Fatalf("status = %s, want error", result.Status)
+			}
+			if result.Output != want {
+				t.Fatalf("output = %q,\n want the recovery message %q", result.Output, want)
+			}
+			if strings.Contains(result.Output, "must be at least 1") {
+				t.Fatalf("still returns the terse minimum error: %q", result.Output)
+			}
+		})
 	}
 }
 

internal/tools/file_tools_test.go

@@ -829,3 +829,61 @@ func TestGlobSkipsWorkspaceExcludedDirectoriesAndBinaryFiles(t *testing.T) {
 		t.Fatalf("expected only keep.txt, got:\n%s", res.Output)
 	}
 }
+
+// A backwards range (end_line < start_line) is a caller-arithmetic slip, not a
+// fatal error: read_file recovers by reading just start_line and surfaces a note
+// explaining the adjustment, matching how an end_line past EOF is auto-clamped.
+// It must never hard-error the way it used to.
+func TestReadFileToolRecoversBackwardsRange(t *testing.T) {
+	root := t.TempDir()
+	writeTestFile(t, filepath.Join(root, "notes.txt"), "alpha\nbeta\ngamma\ndelta\nepsilon")
+
+	result := NewReadFileTool(root).Run(context.Background(), map[string]any{
+		"path":       "notes.txt",
+		"start_line": 3,
+		"end_line":   2, // backwards
+	})
+
+	if result.Status != StatusOK {
+		t.Fatalf("backwards range must recover, not error; got %s: %s", result.Status, result.Output)
+	}
+	if strings.Contains(result.Output, "must be greater than or equal to start_line") {
+		t.Fatalf("backwards range still hard-errors: %q", result.Output)
+	}
+	for _, want := range []string{
+		"end_line 2 was before start_line 3", // the note
+		"only line 3 was read",
+		"3 | gamma",
+	} {
+		if !strings.Contains(result.Output, want) {
+			t.Fatalf("expected %q in output, got %q", want, result.Output)
+		}
+	}
+	// It read exactly start_line, nothing adjacent.
+	if strings.Contains(result.Output, "beta") || strings.Contains(result.Output, "delta") {
+		t.Fatalf("recovery read the wrong lines: %q", result.Output)
+	}
+}
+
+// A valid range with end_line >= start_line is byte-for-byte unchanged: the
+// recovery path must not touch normal reads (no stray note, exact same window).
+func TestReadFileToolValidRangeUnchangedByRecovery(t *testing.T) {
+	root := t.TempDir()
+	writeTestFile(t, filepath.Join(root, "notes.txt"), "alpha\nbeta\ngamma\ndelta\nepsilon")
+
+	result := NewReadFileTool(root).Run(context.Background(), map[string]any{
+		"path":       "notes.txt",
+		"start_line": 2,
+		"end_line":   4,
+	})
+	if result.Status != StatusOK {
+		t.Fatalf("valid range failed: %s: %s", result.Status, result.Output)
+	}
+	// Byte-for-byte: header, blank separator, the exact selected lines, and no
+	// recovery note or trailing content. An altered header/separator or a stray
+	// out-of-range line would fail this where a substring check would not.
+	const want = "File: notes.txt (lines 2-4 of 5)\n\n2 | beta\n3 | gamma\n4 | delta"
+	if result.Output != want {
+		t.Fatalf("valid-range output changed by the recovery path:\n got: %q\nwant: %q", result.Output, want)
+	}
+}

internal/tools/read_file.go

@@ -106,8 +106,18 @@ func renderReadFileRange(absolutePath string, relativePath string, total int, st
 	if endLine == 0 || endLine > total {
 		endLine = total
 	}
+	// A backwards range (end_line < start_line) is a caller slip, not a fatal
+	// error. Recover the way an end_line past EOF is already recovered a few lines
+	// up (clamped down to the last line): clamp end_line UP to start_line and read
+	// just that one line. Erroring here instead cost the caller a whole retry for
+	// an off-by-arithmetic mistake, and it was inconsistent — start_line past EOF
+	// and end_line past EOF both recover gracefully; only this case hard-failed.
+	// The recovery is surfaced in the output (rangeNote) so it is never silent: the
+	// caller sees it read a different range than it asked for and why.
+	rangeNote := ""
 	if endLine < startLine {
-		return errorResult("Error: Invalid arguments for read_file: end_line must be greater than or equal to start_line")
+		rangeNote = fmt.Sprintf("(note: end_line %d was before start_line %d, so only line %d was read — pass end_line >= start_line to read a wider range)", endLine, startLine, startLine)
+		endLine = startLine
 	}
 
 	truncated := false
@@ -135,7 +145,12 @@ func renderReadFileRange(absolutePath string, relativePath string, total int, st
 		budgetedOutput = newHeadTailOutputBudgetBuilder(readOutputBudgetBytes, "use start_line/end_line or max_lines to continue with a smaller range")
 	}
 	budgetedOutput.WriteString(header)
-	budgetedOutput.WriteString("\n\n")
+	budgetedOutput.WriteString("\n")
+	if rangeNote != "" {
+		budgetedOutput.WriteString(rangeNote)
+		budgetedOutput.WriteString("\n")
+	}
+	budgetedOutput.WriteString("\n")
 	if err := appendReadFileRange(budgetedOutput, absolutePath, startLine, selectedLines, width); err != nil {
 		return errorResult("Error reading file " + relativePath + ": " + err.Error())
 	}

internal/tools/registry.go

@@ -2,6 +2,7 @@ package tools
 
 import (
 	"context"
+	"sort"
 
 	"github.com/Gitlawb/zero/internal/redaction"
 	"github.com/Gitlawb/zero/internal/sandbox"
@@ -113,6 +114,9 @@ func (registry *Registry) All() []Tool {
 	for _, tool := range registry.tools {
 		tools = append(tools, tool)
 	}
+	sort.Slice(tools, func(left, right int) bool {
+		return tools[left].Name() < tools[right].Name()
+	})
 	return tools
 }
 

internal/tools/registry_test.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"os"
 	"path/filepath"
+	"slices"
 	"strings"
 	"testing"
 
@@ -72,6 +73,23 @@ func TestCoreReadOnlyToolsExposeSafeMetadata(t *testing.T) {
 	}
 }
 
+func TestRegistryAllReturnsToolsSortedByName(t *testing.T) {
+	registry := NewRegistry()
+	for _, name := range []string{"write_file", "bash", "read_file"} {
+		registry.Register(fakePlainTool{baseTool: baseTool{name: name, description: name}})
+	}
+
+	all := registry.All()
+	got := make([]string, 0, len(all))
+	for _, tool := range all {
+		got = append(got, tool.Name())
+	}
+	want := []string{"bash", "read_file", "write_file"}
+	if !slices.Equal(got, want) {
+		t.Fatalf("Registry.All() order = %v, want %v", got, want)
+	}
+}
+
 func TestCoreNetworkToolsExposeSafetyMetadata(t *testing.T) {
 	// web_search is only registered when a backend is configured.
 	t.Setenv("ZERO_WEBSEARCH_BASE_URL", "https://search.example/api")

internal/trace/emit.go

@@ -106,6 +106,7 @@ func WriteNDJSON(w io.Writer, t *TurnTrace) error {
 	for _, p := range t.PrefixHashes {
 		if err := enc.Encode(map[string]any{
 			"type":                "prefix_hash",
+			"system_prompt":       p.SystemPromptHash,
 			"base_instructions":   p.BaseInstructionsHash,
 			"confirmation_policy": p.ConfirmationPolicyHash,
 			"project_context":     p.ProjectContextHash,
@@ -136,6 +137,27 @@ func WriteNDJSON(w io.Writer, t *TurnTrace) error {
 			return err
 		}
 	}
+	for _, event := range t.TaskStates {
+		if err := enc.Encode(map[string]any{
+			"type":                 "task_state",
+			"revision":             event.Revision,
+			"status":               event.Status,
+			"plan_pending":         event.PlanPending,
+			"plan_in_progress":     event.PlanInProgress,
+			"plan_completed":       event.PlanCompleted,
+			"plan_failed":          event.PlanFailed,
+			"tools_succeeded":      event.ToolsSucceeded,
+			"tools_failed":         event.ToolsFailed,
+			"verification_passed":  event.VerificationPassed,
+			"verification_failed":  event.VerificationFailed,
+			"verification_outcome": event.VerificationOutcome,
+			"changed_file_count":   event.ChangedFileCount,
+			"completion_decision":  event.CompletionDecision,
+			"plan_parity":          event.PlanParity,
+		}); err != nil {
+			return err
+		}
+	}
 	return nil
 }
 
@@ -205,6 +227,15 @@ func WriteText(w io.Writer, t *TurnTrace) error {
 				event.Truncated, event.Reason, event.SpillCreated)
 		}
 	}
+	if len(t.TaskStates) > 0 {
+		latest := t.TaskStates[len(t.TaskStates)-1]
+		write("task state: revision=%d status=%s plan=%d/%d/%d/%d tools=%d/%d verification=%d/%d(%s) files=%d plan_parity=%s completion=%s\n",
+			latest.Revision, latest.Status, latest.PlanPending, latest.PlanInProgress,
+			latest.PlanCompleted, latest.PlanFailed, latest.ToolsSucceeded,
+			latest.ToolsFailed, latest.VerificationPassed, latest.VerificationFailed,
+			latest.VerificationOutcome, latest.ChangedFileCount, latest.PlanParity,
+			latest.CompletionDecision)
+	}
 	return firstErr
 }
 

internal/trace/parse.go

@@ -101,13 +101,14 @@ func ReadNDJSON(r io.Reader) (*TurnTrace, error) {
 			if !sawTraceHeader {
 				return nil, errors.New("parse trace: not a valid NDJSON trace (no type:trace header)")
 			}
-			// Round-trip the seven prefix-hash fields. Missing fields are
+			// Round-trip the prefix-hash fields. Missing fields are
 			// accepted as empty strings (a partially-written trace from a
 			// crashed emitter must not fatal a parse). The decoder tolerates
-			// any shape the encoder produced and the seven keys are the
+			// any shape the encoder produced and the keys are the
 			// contract — adding a new field is non-breaking; renaming one
 			// requires a schema version bump.
 			t.PrefixHashes = append(t.PrefixHashes, PrefixHash{
+				SystemPromptHash:       stringField(obj, "system_prompt"),
 				BaseInstructionsHash:   stringField(obj, "base_instructions"),
 				ConfirmationPolicyHash: stringField(obj, "confirmation_policy"),
 				ProjectContextHash:     stringField(obj, "project_context"),
@@ -131,6 +132,26 @@ func ReadNDJSON(r io.Reader) (*TurnTrace, error) {
 				Reason:                  stringField(obj, "reason"),
 				SpillCreated:            boolField(obj, "spill_created"),
 			})
+		case "task_state":
+			if !sawTraceHeader {
+				return nil, errors.New("parse trace: not a valid NDJSON trace (no type:trace header)")
+			}
+			t.TaskStates = append(t.TaskStates, TaskStateEvent{
+				Revision:            int(parseInt64(obj["revision"])),
+				Status:              stringField(obj, "status"),
+				PlanPending:         int(parseInt64(obj["plan_pending"])),
+				PlanInProgress:      int(parseInt64(obj["plan_in_progress"])),
+				PlanCompleted:       int(parseInt64(obj["plan_completed"])),
+				PlanFailed:          int(parseInt64(obj["plan_failed"])),
+				ToolsSucceeded:      int(parseInt64(obj["tools_succeeded"])),
+				ToolsFailed:         int(parseInt64(obj["tools_failed"])),
+				VerificationPassed:  int(parseInt64(obj["verification_passed"])),
+				VerificationFailed:  int(parseInt64(obj["verification_failed"])),
+				VerificationOutcome: stringField(obj, "verification_outcome"),
+				ChangedFileCount:    int(parseInt64(obj["changed_file_count"])),
+				CompletionDecision:  stringField(obj, "completion_decision"),
+				PlanParity:          stringField(obj, "plan_parity"),
+			})
 		default:
 			// Unknown event type: tolerate (forward-compat) but only after a
 			// header has been seen.
@@ -151,8 +172,8 @@ func ReadNDJSON(r io.Reader) (*TurnTrace, error) {
 	if !sawTraceHeader {
 		return nil, errors.New("parse trace: non-empty input had no type:trace header")
 	}
-	if len(t.Spans) == 0 && len(t.Counters) == 0 && len(t.PrefixHashes) == 0 && len(t.OutputBudgets) == 0 {
-		return nil, errors.New("parse trace: header present but no spans, counters, or prefix hashes recovered (corrupt or truncated)")
+	if len(t.Spans) == 0 && len(t.Counters) == 0 && len(t.PrefixHashes) == 0 && len(t.OutputBudgets) == 0 && len(t.TaskStates) == 0 {
+		return nil, errors.New("parse trace: header present but no events recovered (corrupt or truncated)")
 	}
 	return t, nil
 }

internal/trace/recorder.go

@@ -29,6 +29,10 @@ type Recorder struct {
 	firstActionStamped  bool
 }
 
+// maxTaskStateEvents bounds trace growth in pathological long runs. Once full,
+// the latest aggregate replaces the tail so final state is never lost.
+const maxTaskStateEvents = 128
+
 // NewRecorder returns a ready recorder. sessionID correlates with the agent
 // session (Options.SessionID); runID is a per-Run sequence; profile is an
 // optional label (e.g. "cold", "warm") for benchmark runs.
@@ -203,6 +207,24 @@ func (r *Recorder) EmitOutputBudget(event OutputBudgetEvent) {
 	r.tr.OutputBudgets = append(r.tr.OutputBudgets, event)
 }
 
+// EmitTaskState records one content-free structured task snapshot. Calls retain
+// event order so revisions can be replayed alongside the tool and turn stream.
+func (r *Recorder) EmitTaskState(event TaskStateEvent) {
+	if r == nil {
+		return
+	}
+	r.mu.Lock()
+	defer r.mu.Unlock()
+	if r.finished {
+		return
+	}
+	if len(r.tr.TaskStates) < maxTaskStateEvents {
+		r.tr.TaskStates = append(r.tr.TaskStates, event)
+		return
+	}
+	r.tr.TaskStates[len(r.tr.TaskStates)-1] = event
+}
+
 // Finish stamps CompletedAt, derives each span's parent (by interval
 // containment) and exclusive time, and returns a snapshot of the trace. Calling
 // Finish more than once returns the same snapshot.
@@ -223,6 +245,7 @@ func (r *Recorder) Finish() *TurnTrace {
 	snap.Counters = append([]Counter(nil), r.tr.Counters...)
 	snap.PrefixHashes = append([]PrefixHash(nil), r.tr.PrefixHashes...)
 	snap.OutputBudgets = append([]OutputBudgetEvent(nil), r.tr.OutputBudgets...)
+	snap.TaskStates = append([]TaskStateEvent(nil), r.tr.TaskStates...)
 	return &snap
 }
 

internal/trace/task_state_test.go

@@ -0,0 +1,66 @@
+package trace
+
+import (
+	"bytes"
+	"strings"
+	"testing"
+)
+
+func TestTaskStateTraceRoundTripContainsNoTaskContent(t *testing.T) {
+	recorder := NewRecorder("session", "run", "")
+	recorder.Start()
+	event := TaskStateEvent{
+		Revision:            4,
+		Status:              "active",
+		PlanPending:         2,
+		PlanCompleted:       1,
+		ToolsSucceeded:      3,
+		ToolsFailed:         1,
+		VerificationPassed:  1,
+		VerificationOutcome: "passed",
+		ChangedFileCount:    2,
+		PlanParity:          "match",
+	}
+	recorder.EmitTaskState(event)
+
+	var output bytes.Buffer
+	if err := WriteNDJSON(&output, recorder.Finish()); err != nil {
+		t.Fatalf("WriteNDJSON: %v", err)
+	}
+	if strings.Contains(output.String(), "secret objective") || strings.Contains(output.String(), "private.go") {
+		t.Fatalf("task content leaked into trace: %s", output.String())
+	}
+	if strings.Contains(output.String(), "objective_hash") {
+		t.Fatalf("trace contains a confirmable objective fingerprint: %s", output.String())
+	}
+	parsed, err := ReadNDJSON(strings.NewReader(output.String()))
+	if err != nil {
+		t.Fatalf("ReadNDJSON: %v", err)
+	}
+	if len(parsed.TaskStates) != 1 || parsed.TaskStates[0] != event {
+		t.Fatalf("unexpected round trip: %#v", parsed.TaskStates)
+	}
+}
+
+func TestTaskStateIsDocumentedAsOptionalTraceEvent(t *testing.T) {
+	for _, key := range OptionalEventKeys() {
+		if key == "task_state" {
+			return
+		}
+	}
+	t.Fatal("task_state missing from OptionalEventKeys")
+}
+
+func TestTaskStateTraceIsBoundedAndRetainsLatestSnapshot(t *testing.T) {
+	recorder := NewRecorder("session", "run", "")
+	for revision := 1; revision <= maxTaskStateEvents+20; revision++ {
+		recorder.EmitTaskState(TaskStateEvent{Revision: revision})
+	}
+	events := recorder.Finish().TaskStates
+	if len(events) != maxTaskStateEvents {
+		t.Fatalf("task state events = %d, want cap %d", len(events), maxTaskStateEvents)
+	}
+	if got := events[len(events)-1].Revision; got != maxTaskStateEvents+20 {
+		t.Fatalf("latest revision = %d, want %d", got, maxTaskStateEvents+20)
+	}
+}

internal/trace/trace.go

@@ -32,6 +32,7 @@ const (
 	SpanToolPartition   = "tool_partition"   // partitioning the tool set for the prompt
 	SpanProviderConnect = "provider_connect" // client.Do in the provider seam
 	SpanProviderQueue   = "provider_queue"   // pre-send OAuth/token resolve
+	SpanProviderPrewarm = "provider_prewarm" // best-effort turn-session prewarm probe
 	SpanGeneration      = "generation"       // streaming a model completion
 	SpanToolExecution   = "tool_execution"   // executing tool calls
 	SpanPermissionWait  = "permission_wait"  // waiting on a permission prompt
@@ -53,6 +54,14 @@ const (
 	CounterInputTokens       = "input_tokens"
 	CounterCachedInputTokens = "cached_input_tokens"
 	CounterOutputTokens      = "output_tokens"
+	CounterPrewarmAttempts   = "prewarm_attempts"
+	CounterPrefixStable      = "prefix_stable"
+	CounterPrefixDrift       = "prefix_drift"
+	// CounterPostureEscalations counts one-shot in-run posture escalations by
+	// the execution-profile controller. Deliberately distinct from
+	// model_switches: a posture escalation changes loop policy knobs, never the
+	// model or session.
+	CounterPostureEscalations = "posture_escalations"
 )
 
 // Span is one named wall interval attributed to part of a run. Each stamp is
@@ -94,6 +103,27 @@ type OutputBudgetEvent struct {
 	SpillCreated            bool   `json:"spill_created"`
 }
 
+// TaskStateEvent is a content-free snapshot of the run's structured task
+// projection. Objective text, plan content, tool output, file paths, and
+// confirmable content fingerprints are deliberately excluded so tracing cannot
+// become a secret-bearing side channel.
+type TaskStateEvent struct {
+	Revision            int    `json:"revision"`
+	Status              string `json:"status"`
+	PlanPending         int    `json:"plan_pending"`
+	PlanInProgress      int    `json:"plan_in_progress"`
+	PlanCompleted       int    `json:"plan_completed"`
+	PlanFailed          int    `json:"plan_failed"`
+	ToolsSucceeded      int    `json:"tools_succeeded"`
+	ToolsFailed         int    `json:"tools_failed"`
+	VerificationPassed  int    `json:"verification_passed"`
+	VerificationFailed  int    `json:"verification_failed"`
+	VerificationOutcome string `json:"verification_outcome,omitempty"`
+	ChangedFileCount    int    `json:"changed_file_count"`
+	CompletionDecision  string `json:"completion_decision,omitempty"`
+	PlanParity          string `json:"plan_parity"`
+}
+
 // TurnTrace is the finished record for one agent.Run. It is the value
 // emitters serialize; it is not mutated after Finish returns a snapshot.
 type TurnTrace struct {
@@ -109,34 +139,19 @@ type TurnTrace struct {
 	Counters            []Counter           `json:"counters"`
 	PrefixHashes        []PrefixHash        `json:"prefix_hashes,omitempty"`
 	OutputBudgets       []OutputBudgetEvent `json:"output_budgets,omitempty"`
+	TaskStates          []TaskStateEvent    `json:"task_states,omitempty"`
 }
 
 // PrefixHash is one fingerprint of the prompt prefix emitted by an agent run.
-// The seven fields decompose the cacheable part of the request so a downstream
-// observer can detect which sub-component drifted turn-over-turn:
-// base_instructions (the embedded core system prompt), confirmation_policy,
-// project_context (AGENTS.md / ZERO.md chain), skills, tools, schema (tool
-// JSON schemas), and complete_prefix (SHA-256 of the canonical concatenation
-// of the other six). All hashes are hex-encoded SHA-256.
-//
-// Scope: the fingerprint covers 4 of 11 sections of buildSystemPrompt. The
-// seven sections NOT covered — modelPromptAddendum, sessionRuntimeContext,
-// approvedCommandPrefixContext, workspaceSeedContext, userGuidelines,
-// specialistDelegationContext, responseStyleContext — fire only for
-// non-default Options. For default Options (the common case), the four
-// captured substrings are the full prompt and the fingerprint is accurate.
-//
-// A run with a stable CompletePrefix across turns means the four captured
-// sub-components are byte-identical. A run where CompletePrefix changes
-// names the captured sub-component that drifted, but does NOT rule out
-// drift in the seven uncaptured sections. modelPromptAddendum in particular
-// changes on a model switch (a model_switches counter the trace already
-// emits), so a consumer correlating CompletePrefix stability with
-// cached_input_tokens must cross-check the model_switches counter to
-// disambiguate "no drift" from "drift in an uncaptured section." The fields
-// are emitted as a "prefix_hash" event in the NDJSON trace (see
-// WriteNDJSON).
+// SystemPromptHash covers the exact joined system-message content, including
+// every dynamic section. ToolsHash covers tool names and descriptions in their
+// emitted order; SchemaHash covers their canonical parameter schemas in that
+// same order. CompletePrefixHash combines those three provider-visible parts.
+// The remaining hashes retain narrower diagnostic boundaries so a trace can
+// identify common sources of system-prompt drift. All hashes are hex-encoded
+// SHA-256 and emitted as a "prefix_hash" NDJSON event.
 type PrefixHash struct {
+	SystemPromptHash       string `json:"system_prompt"`
 	BaseInstructionsHash   string `json:"base_instructions"`
 	ConfirmationPolicyHash string `json:"confirmation_policy"`
 	ProjectContextHash     string `json:"project_context"`
@@ -281,6 +296,12 @@ func OptionalEventKeys() []string {
 		"counter:" + CounterAcceptanceChecks,
 		"counter:" + CounterPollingTurn,
 		"counter:" + CounterModelSwitches,
+		"span:" + SpanProviderPrewarm,
+		"counter:" + CounterPrewarmAttempts,
+		"counter:" + CounterPrefixStable,
+		"counter:" + CounterPrefixDrift,
+		"counter:" + CounterPostureEscalations,
 		"event:prefix_hash",
+		"task_state",
 	}
 }

internal/trace/trace_test.go

@@ -374,6 +374,7 @@ func TestReadNDJSONRoundTrip(t *testing.T) {
 	// event type. Insertion order is preserved by the parser (no sort
 	// is applied on read or write).
 	r.EmitPrefixHash(PrefixHash{
+		SystemPromptHash:       "system1",
 		BaseInstructionsHash:   "b1",
 		ConfirmationPolicyHash: "c1",
 		ProjectContextHash:     "p1",
@@ -383,6 +384,7 @@ func TestReadNDJSONRoundTrip(t *testing.T) {
 		CompletePrefixHash:     "complete1",
 	})
 	r.EmitPrefixHash(PrefixHash{
+		SystemPromptHash:       "system2",
 		BaseInstructionsHash:   "b2",
 		ConfirmationPolicyHash: "c2",
 		ProjectContextHash:     "p2",
@@ -427,10 +429,10 @@ func TestReadNDJSONRoundTrip(t *testing.T) {
 	if parsed.PrefixHashes[0].CompletePrefixHash != "complete1" || parsed.PrefixHashes[1].CompletePrefixHash != "complete2" {
 		t.Fatalf("prefix_hash insertion order lost: got %+v want [complete1, complete2]", parsed.PrefixHashes)
 	}
-	if parsed.PrefixHashes[0].BaseInstructionsHash != "b1" || parsed.PrefixHashes[0].SchemaHash != "x1" {
+	if parsed.PrefixHashes[0].SystemPromptHash != "system1" || parsed.PrefixHashes[0].BaseInstructionsHash != "b1" || parsed.PrefixHashes[0].SchemaHash != "x1" {
 		t.Fatalf("prefix_hash sub-hashes lost on first event: got %+v", parsed.PrefixHashes[0])
 	}
-	if parsed.PrefixHashes[1].BaseInstructionsHash != "b2" || parsed.PrefixHashes[1].SchemaHash != "x2" {
+	if parsed.PrefixHashes[1].SystemPromptHash != "system2" || parsed.PrefixHashes[1].BaseInstructionsHash != "b2" || parsed.PrefixHashes[1].SchemaHash != "x2" {
 		t.Fatalf("prefix_hash sub-hashes lost on second event: got %+v", parsed.PrefixHashes[1])
 	}
 }
@@ -577,3 +579,12 @@ func TestCounterPrecisionRoundTrip(t *testing.T) {
 		t.Fatalf("counter round-trip = %d, want %d (float64 precision loss)", got, big)
 	}
 }
+
+func TestOptionalEventKeysIncludePostureEscalations(t *testing.T) {
+	for _, key := range OptionalEventKeys() {
+		if key == "counter:"+CounterPostureEscalations {
+			return
+		}
+	}
+	t.Fatalf("posture_escalations missing from OptionalEventKeys: %v", OptionalEventKeys())
+}

internal/tui/btw.go

@@ -0,0 +1,325 @@
+package tui
+
+import (
+	"fmt"
+	"strings"
+
+	tea "charm.land/bubbletea/v2"
+
+	"github.com/Gitlawb/zero/internal/sessions"
+	"github.com/Gitlawb/zero/internal/usage"
+)
+
+const (
+	btwSessionTag = "btw"
+	// btwRunIDGap keeps the hidden parent's current run below the first side run.
+	// A parent cannot start another user turn while its surface is hidden; the
+	// monotonic btwRunIDSeq separately prevents collisions across BTW re-entry.
+	btwRunIDGap = 1_000_000
+
+	btwContextBoundary = `You are in an isolated side conversation. The inherited session history is reference context only.
+Do not continue or complete earlier tasks, plans, tool calls, approvals, or requests. Only instructions submitted after this boundary are active.
+Answer questions and explore lightly. Do not modify workspace state unless the user explicitly asks for that mutation in this side conversation.
+Nothing from this side conversation will be merged into the main session.`
+)
+
+// btwState keeps the main surface alive while an isolated side conversation is
+// visible. The parent model continues receiving its own run messages, but its
+// transcript and session events stay hidden until the user returns.
+type btwState struct {
+	active           bool
+	parent           *model
+	sideRunIDBase    int
+	parentNeedsInput bool
+}
+
+func (m model) handleBTWCommand(question string) (model, tea.Cmd) {
+	question = strings.TrimSpace(question)
+	if m.btw.active {
+		if question != "" {
+			return m.appendSystemNotice("A BTW conversation is already active. Run /btw with no question to return to the main session."), nil
+		}
+		return m.leaveBTW()
+	}
+	if m.compactInFlight {
+		return m.appendSystemNotice("Compaction is running. Wait for it to finish before opening a BTW conversation."), nil
+	}
+	if m.exiting {
+		return m.appendSystemNotice("Zero is waiting for the current run to finish saving before exit. A BTW conversation cannot start now."), nil
+	}
+	if m.activeSession.SessionID == "" {
+		return m.appendSystemNotice("Start the main session with a prompt before opening a BTW conversation."), nil
+	}
+	if m.sessionStore == nil {
+		return m.appendSystemNotice("BTW conversation unavailable: no session store configured."), nil
+	}
+
+	fork, err := m.sessionStore.Fork(m.activeSession.SessionID, sessions.ForkInput{
+		SessionKind: sessions.SessionKindSide,
+		Title:       btwTitle(m.activeSession.Title),
+		Cwd:         m.cwd,
+		ModelID:     m.modelName,
+		Provider:    m.providerName,
+		Tag:         btwSessionTag,
+	})
+	if err != nil {
+		return m.appendSystemNotice("Could not open BTW conversation: " + err.Error()), nil
+	}
+	events, err := m.resumeEvents(fork.SessionID)
+	if err != nil {
+		return m.appendSystemNotice("Could not load BTW conversation: " + err.Error()), nil
+	}
+
+	// Foreground loops are session-scoped. Stop them only after the side session
+	// is ready, so a failed fork cannot destroy work in the main session.
+	parent := m
+	clearedLoops := 0
+	if updated, cleared := parent.clearLoopsForSessionSwitch(); cleared > 0 {
+		parent = updated
+		clearedLoops = cleared
+		parent.transcript = reduceTranscript(parent.transcript, transcriptAction{
+			kind: actionAppendSystem,
+			text: fmt.Sprintf("Stopped %d loop(s) before opening the BTW conversation.", cleared),
+		})
+	}
+	parent.btw = btwState{}
+	// A scrollback print that was already scheduled may acknowledge after the
+	// side surface is active. The hidden model does not receive that unscoped
+	// acknowledgement, so clear its print latch and rebuild on return.
+	parent.printInFlight = false
+	parent.flushQueue = nil
+
+	side := parent
+	side.activeSession = fork
+	side.sessionEvents = events
+	side.transcript = initialTranscript()
+	side.transcript = appendRow(side.transcript, rowSystem, "BTW conversation · isolated from the main session · /btw or Ctrl+C to return")
+	if clearedLoops > 0 {
+		side.transcript = appendRow(side.transcript, rowSystem, fmt.Sprintf("Stopped %d main-session loop(s) before opening this conversation.", clearedLoops))
+	}
+	side.transcript = appendTranscriptRowsDedup(side.transcript, transcriptRowsFromSessionEvents(events))
+	side.printInFlight = false
+	side.flushQueue = nil
+	side.resetFlushFrontier("· btw conversation ·")
+	side.pending = false
+	side.exiting = false
+	side.runCancel = nil
+	side.activeRunID = 0
+	side.runID = maxInt(parent.runID+btwRunIDGap, parent.btwRunIDSeq)
+	side.flushRunIDs = map[int]string{}
+	side.liveUsageCounts = map[int]int{}
+	side.usageTracker = usage.NewTracker(usage.TrackerOptions{Registry: &side.modelCatalog, Now: side.now})
+	side.lastUsage = usage.Normalized{}
+	side.lastUsageSeen = false
+	side.unpricedRequests = 0
+	side.unpricedTokens = 0
+	side.compactRequests = 0
+	side.compactFrame = 0
+	side.lastCompactResult = nil
+	side.lastCompactError = ""
+	side.turnLatencySum = 0
+	side.turnLatencyCount = 0
+	side.turnTTFTSum = 0
+	side.turnTTFTCount = 0
+	side.pendingPermission = nil
+	side.pendingAskUser = nil
+	side.pendingSpecReview = nil
+	side.queuedMessage = ""
+	side.lastPrompt = ""
+	side.lastImages = nil
+	side.lastImageLabels = nil
+	side.lastDocuments = nil
+	side.inputHistory = nil
+	side.historyIdx = 0
+	side.historyDraft = ""
+	if question == "" {
+		side.pendingImages = nil
+		side.pendingImageLabels = nil
+		side.pendingDocuments = nil
+	}
+	side.loops = nil
+	side.activeLoopID = ""
+	side.loopTicking = false
+	side.specialists.clear()
+	side.plan.clear()
+	side.planDetailGen++
+	side.streamingText = nil
+	side.streamingReasoning = ""
+	side.streamingReasoningExpanded = false
+	side.clearStreamingToolCall()
+	side.resetStreamingFade()
+	side.btw = btwState{
+		active:        true,
+		parent:        &parent,
+		sideRunIDBase: side.runID,
+	}
+
+	if question == "" {
+		return side, nil
+	}
+	return side.launchPrompt(question)
+}
+
+func (m model) leaveBTW() (model, tea.Cmd) {
+	if !m.btw.active || m.btw.parent == nil {
+		return m, nil
+	}
+	if m.pending {
+		return m.appendSystemNotice("A BTW response is still running. Press Esc twice to cancel it, or wait for it to finish before returning."), nil
+	}
+	if len(m.flushRunIDs) > 0 {
+		return m.appendSystemNotice("The cancelled BTW response is still saving its session events. Wait for it to finish before returning."), nil
+	}
+	if m.compactInFlight {
+		return m.appendSystemNotice("BTW compaction is still running. Wait for it to finish before returning."), nil
+	}
+	m, _ = m.clearLoopsForSessionSwitch()
+	parent := *m.btw.parent
+	parent.btwRunIDSeq = maxInt(parent.btwRunIDSeq, m.runID)
+	parent.btw = btwState{}
+	// A hidden parent completion may have scheduled an unscoped git-sweep result
+	// that landed on the side surface. Re-run it after restoring the parent so
+	// its FILES sidebar cannot stay permanently marked in-flight or stale.
+	parent.gitSweepInFlight = false
+	var sweepCmd tea.Cmd
+	parent, sweepCmd = parent.maybeGitSweep()
+	// The side surface owns its own spinner tick chain. If the hidden parent is
+	// still active, explicitly restart that chain after restoring it.
+	parent.spinnerTicking = false
+	var spinnerCmd tea.Cmd
+	if parent.pending || parent.compactInFlight || parent.doctorInFlight || parent.sidebarHasAgents() || parent.aimlapiOnboardAnimating() {
+		parent.spinnerTicking = true
+		spinnerCmd = parent.spinner.Tick
+	}
+	parent.transcript = reduceTranscript(parent.transcript, transcriptAction{
+		kind: actionAppendSystem,
+		text: "Returned from the isolated BTW conversation. Its messages were not added to this session.",
+	})
+	parent.resetFlushFrontier("· returned from btw ·")
+	return parent, batchCommands(sweepCmd, spinnerCmd)
+}
+
+func btwCommandUnavailable(command parsedCommand) bool {
+	arg := strings.ToLower(strings.TrimSpace(command.text))
+	switch command.kind {
+	case commandNew, commandResume, commandRetitle, commandSpec, commandLoop,
+		commandRewind, commandCompact, commandSTTModel, commandMCP:
+		return true
+	case commandModel:
+		return arg != "list" && arg != "ls"
+	case commandProvider:
+		return arg != "status"
+	case commandTurns, commandProfile:
+		return arg != "" && arg != "status"
+	case commandTheme:
+		return arg != "list"
+	case commandConfig:
+		return arg != ""
+	default:
+		return false
+	}
+}
+
+// resizeBTWParent carries terminal geometry into the hidden parent. Window-size
+// messages have no run ID, so normal BTW routing intentionally leaves them on
+// the visible side surface; copying the layout fields prevents stale wrapping
+// after the parent is restored.
+func (m model) resizeBTWParent(msg tea.WindowSizeMsg) model {
+	if !m.btw.active || m.btw.parent == nil {
+		return m
+	}
+	parent := *m.btw.parent
+	parent.width = msg.Width
+	parent.height = msg.Height
+	parent = parent.clearHover()
+	parent.lineAges = nil
+	parent.lastStreamActivity = parent.now()
+	parent.input.SetWidth(maxInt(20, chatWidth(msg.Width)-14))
+	m.btw.parent = &parent
+	return m
+}
+
+func btwTitle(parent string) string {
+	parent = strings.TrimSpace(parent)
+	if parent == "" {
+		return "BTW conversation"
+	}
+	return parent + " (BTW)"
+}
+
+// routeBTWParentMessage lets a main-session run finish while the side surface is
+// active. Side run IDs start far above the captured parent's counter, so messages
+// below that boundary unambiguously belong to the hidden parent model.
+func (m model) routeBTWParentMessage(msg tea.Msg) (model, tea.Cmd, bool) {
+	if !m.btw.active || m.btw.parent == nil {
+		return m, nil, false
+	}
+	if title, ok := msg.(sessionTitleGeneratedMsg); ok {
+		if title.sessionID != m.btw.parent.activeSession.SessionID {
+			return m, nil, false
+		}
+		return m.routeBTWMessageToParent(msg)
+	}
+	runID, ok := btwMessageRunID(msg)
+	if !ok || runID <= 0 || runID >= m.btw.sideRunIDBase {
+		return m, nil, false
+	}
+	return m.routeBTWMessageToParent(msg)
+}
+
+func (m model) routeBTWMessageToParent(msg tea.Msg) (model, tea.Cmd, bool) {
+	parentNext, cmd := m.btw.parent.updateModel(msg)
+	parent, ok := parentNext.(model)
+	if !ok {
+		return m, cmd, true
+	}
+	parent.btw = btwState{}
+	m.btw.parent = &parent
+	switch msg.(type) {
+	case permissionRequestMsg, askUserRequestMsg:
+		if !m.btw.parentNeedsInput {
+			m.btw.parentNeedsInput = true
+			m = m.appendSystemNotice("The main session needs your input. Return with /btw or Ctrl+C to respond.")
+		}
+	case agentResponseMsg:
+		m.btw.parentNeedsInput = parent.pendingPermission != nil || parent.pendingAskUser != nil
+	}
+	return m, cmd, true
+}
+
+func btwMessageRunID(msg tea.Msg) (int, bool) {
+	switch typed := msg.(type) {
+	case agentTextMsg:
+		return typed.runID, true
+	case agentReasoningMsg:
+		return typed.runID, true
+	case agentUsageMsg:
+		return typed.runID, true
+	case agentResponseMsg:
+		return typed.runID, true
+	case agentRowMsg:
+		return typed.runID, true
+	case toolCallStreamStartMsg:
+		return typed.runID, true
+	case toolCallStreamDeltaMsg:
+		return typed.runID, true
+	case planUpdateMsg:
+		return typed.runID, true
+	case specialistStartMsg:
+		return typed.runID, true
+	case specialistCompleteMsg:
+		return typed.runID, true
+	case specialistProgressMsg:
+		return typed.runID, true
+	case swarmSessionsMsg:
+		return typed.runID, true
+	case permissionRequestMsg:
+		return typed.runID, true
+	case askUserRequestMsg:
+		return typed.runID, true
+	case recapGeneratedMsg:
+		return typed.runID, true
+	default:
+		return 0, false
+	}
+}

internal/tui/btw_test.go

@@ -0,0 +1,430 @@
+package tui
+
+import (
+	"context"
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+
+	tea "charm.land/bubbletea/v2"
+
+	"github.com/Gitlawb/zero/internal/sessions"
+)
+
+func newBTWTestModel(t *testing.T) model {
+	t.Helper()
+	store := sessions.NewStore(sessions.StoreOptions{RootDir: t.TempDir()})
+	parent, err := store.Create(sessions.CreateInput{
+		SessionID: "main-session",
+		Title:     "Main task",
+		Cwd:       "/repo",
+		ModelID:   "test-model",
+		Provider:  "test-provider",
+	})
+	if err != nil {
+		t.Fatalf("create parent session: %v", err)
+	}
+	for _, payload := range []map[string]any{
+		{"role": "user", "content": "implement the main task"},
+		{"role": "assistant", "content": "working on it"},
+	} {
+		if _, err := store.AppendEvent(parent.SessionID, sessions.AppendEventInput{Type: sessions.EventMessage, Payload: payload}); err != nil {
+			t.Fatalf("append parent event: %v", err)
+		}
+	}
+	loaded, err := store.Get(parent.SessionID)
+	if err != nil || loaded == nil {
+		t.Fatalf("load parent session: session=%#v err=%v", loaded, err)
+	}
+	events, err := store.ReadEvents(parent.SessionID)
+	if err != nil {
+		t.Fatalf("read parent events: %v", err)
+	}
+	m := newModel(context.Background(), Options{
+		Cwd:          "/repo",
+		ModelName:    "test-model",
+		ProviderName: "test-provider",
+		SessionStore: store,
+	})
+	m.activeSession = *loaded
+	m.sessionEvents = events
+	m.transcript = appendTranscriptRowsDedup(initialTranscript(), transcriptRowsFromSessionEvents(events))
+	return m
+}
+
+func TestBTWCommandParsesInlineQuestion(t *testing.T) {
+	got := parseCommand("/btw double-check the approach")
+	if got.kind != commandBTW || got.text != "double-check the approach" {
+		t.Fatalf("parseCommand(/btw ...) = %#v", got)
+	}
+}
+
+func TestBTWCreatesIsolatedForkAndReturnsWithoutMerge(t *testing.T) {
+	m := newBTWTestModel(t)
+	parentID := m.activeSession.SessionID
+	parentEventCount := len(m.sessionEvents)
+
+	side, cmd := m.handleBTWCommand("")
+	if cmd != nil {
+		t.Fatal("bare /btw should not start an agent run")
+	}
+	if !side.btw.active || side.btw.parent == nil {
+		t.Fatal("expected active BTW state with saved parent")
+	}
+	if side.activeSession.SessionKind != sessions.SessionKindSide || side.activeSession.ParentSessionID != parentID {
+		t.Fatalf("side metadata = %#v", side.activeSession)
+	}
+	if side.activeSession.Tag != btwSessionTag {
+		t.Fatalf("side tag = %q, want %q", side.activeSession.Tag, btwSessionTag)
+	}
+	if len(side.sessionEvents) <= parentEventCount {
+		t.Fatalf("side should contain copied context plus fork marker: got %d events", len(side.sessionEvents))
+	}
+	if !strings.Contains(side.sessionPrompt("question"), btwContextBoundary) {
+		t.Fatal("side prompt is missing the inherited-context boundary")
+	}
+
+	updated, err := side.appendSessionEvent(sessions.EventMessage, map[string]any{
+		"role": "assistant", "content": "side-only answer",
+	})
+	if err != nil {
+		t.Fatalf("append side event: %v", err)
+	}
+	returned, _ := updated.leaveBTW()
+	if returned.activeSession.SessionID != parentID {
+		t.Fatalf("returned session = %q, want %q", returned.activeSession.SessionID, parentID)
+	}
+	if len(returned.sessionEvents) != parentEventCount {
+		t.Fatalf("side events merged into parent: got %d events, want %d", len(returned.sessionEvents), parentEventCount)
+	}
+	parentEvents, err := returned.sessionStore.ReadEvents(parentID)
+	if err != nil {
+		t.Fatalf("read parent after return: %v", err)
+	}
+	for _, event := range parentEvents {
+		if strings.Contains(string(event.Payload), "side-only answer") {
+			t.Fatal("side-only event was persisted into the parent session")
+		}
+	}
+}
+
+func TestBTWCanOpenWhileParentRunContinues(t *testing.T) {
+	m := newBTWTestModel(t)
+	m.pending = true
+	m.runID = 7
+	m.activeRunID = 7
+
+	side, _ := m.handleBTWCommand("")
+	if side.pending || side.activeRunID != 0 {
+		t.Fatalf("side inherited parent run state: pending=%v activeRunID=%d", side.pending, side.activeRunID)
+	}
+	if side.btw.parent == nil || !side.btw.parent.pending || side.btw.parent.activeRunID != 7 {
+		t.Fatalf("parent run was not preserved: %#v", side.btw.parent)
+	}
+
+	routed, _, ok := side.routeBTWParentMessage(agentTextMsg{runID: 7, delta: "main progress"})
+	if !ok {
+		t.Fatal("parent run message was not routed")
+	}
+	if strings.Contains(routed.streamingTextString(), "main progress") {
+		t.Fatal("parent streaming output leaked into the side transcript")
+	}
+	if routed.btw.parent == nil || !strings.Contains(routed.btw.parent.streamingTextString(), "main progress") {
+		t.Fatal("parent streaming output was not retained on the hidden parent")
+	}
+}
+
+func TestBTWInlineQuestionStartsSideRun(t *testing.T) {
+	m := newBTWTestModel(t)
+	m.provider = &fakeProvider{}
+
+	side, cmd := m.handleBTWCommand("double-check this assumption")
+	if cmd == nil || !side.pending {
+		t.Fatalf("inline /btw question did not start a run: pending=%v cmd=%v", side.pending, cmd)
+	}
+	if side.lastPrompt != "double-check this assumption" {
+		t.Fatalf("side last prompt = %q", side.lastPrompt)
+	}
+	if side.btw.parent == nil || side.btw.parent.pending {
+		t.Fatal("idle parent should remain idle while the side run starts")
+	}
+	parentEvents, err := side.sessionStore.ReadEvents(side.btw.parent.activeSession.SessionID)
+	if err != nil {
+		t.Fatalf("read parent events: %v", err)
+	}
+	for _, event := range parentEvents {
+		if strings.Contains(string(event.Payload), "double-check this assumption") {
+			t.Fatal("inline side question was written into the parent session")
+		}
+	}
+}
+
+func TestBTWRejectsBeforeMainSessionStarts(t *testing.T) {
+	m := newModel(context.Background(), Options{SessionStore: sessions.NewStore(sessions.StoreOptions{RootDir: t.TempDir()})})
+	got, cmd := m.handleBTWCommand("")
+	if cmd != nil {
+		t.Fatal("pre-session /btw should not start work")
+	}
+	if got.btw.active || !transcriptContains(got.transcript, "Start the main session") {
+		t.Fatalf("unexpected pre-session /btw result: active=%v transcript=%#v", got.btw.active, got.transcript)
+	}
+}
+
+func TestBTWBlocksCommandsThatWouldReplaceItsSession(t *testing.T) {
+	m := newBTWTestModel(t)
+	side, _ := m.handleBTWCommand("")
+	updated, _ := side.dispatchCommand(parseCommand("/new"))
+	got := updated.(model)
+	if !got.btw.active || got.activeSession.SessionKind != sessions.SessionKindSide {
+		t.Fatalf("/new replaced the active BTW session: active=%v metadata=%#v", got.btw.active, got.activeSession)
+	}
+	if !transcriptContains(got.transcript, "unavailable in a BTW conversation") {
+		t.Fatalf("missing blocked-command guidance: %#v", got.transcript)
+	}
+}
+
+func TestBTWBlocksPersistentConfigurationCommands(t *testing.T) {
+	for _, input := range []string{
+		"/model other",
+		"/provider add",
+		"/turns 100",
+		"/profile fast",
+		"/theme dark",
+		"/config recaps off",
+		"/stt-model",
+		"/mcp",
+		"/rewind",
+		"/compact",
+	} {
+		t.Run(input, func(t *testing.T) {
+			m := newBTWTestModel(t)
+			side, _ := m.handleBTWCommand("")
+			updated, _ := side.dispatchCommand(parseCommand(input))
+			got := updated.(model)
+			if !got.btw.active || got.activeSession.SessionKind != sessions.SessionKindSide {
+				t.Fatalf("%s escaped the active BTW session: active=%v metadata=%#v", input, got.btw.active, got.activeSession)
+			}
+			if !transcriptContains(got.transcript, "unavailable in a BTW conversation") {
+				t.Fatalf("%s missing blocked-command guidance: %#v", input, got.transcript)
+			}
+		})
+	}
+}
+
+func TestBTWAllowsReadOnlyConfigurationCommands(t *testing.T) {
+	for _, input := range []string{
+		"/model list",
+		"/provider status",
+		"/turns",
+		"/profile status",
+		"/theme list",
+		"/config",
+	} {
+		t.Run(input, func(t *testing.T) {
+			m := newBTWTestModel(t)
+			side, _ := m.handleBTWCommand("")
+			updated, _ := side.dispatchCommand(parseCommand(input))
+			got := updated.(model)
+			if transcriptContains(got.transcript, "unavailable in a BTW conversation") {
+				t.Fatalf("%s was blocked even though it is read-only", input)
+			}
+		})
+	}
+}
+
+func TestBTWExitBlockedWhileParentRunActive(t *testing.T) {
+	m := newBTWTestModel(t)
+	m.pending = true
+	m.activeRunID = 7
+	side, _ := m.handleBTWCommand("")
+
+	updated, cmd := side.dispatchCommand(parseCommand("/exit"))
+	got := updated.(model)
+	if cmd != nil || got.exiting || !got.btw.active {
+		t.Fatalf("/exit escaped BTW while parent was active: cmd=%v exiting=%v active=%v", cmd, got.exiting, got.btw.active)
+	}
+	if got.btw.parent == nil || !got.btw.parent.pending {
+		t.Fatalf("hidden parent run was not preserved: %#v", got.btw.parent)
+	}
+	if !transcriptContains(got.transcript, "main session is still running") {
+		t.Fatalf("missing active-parent exit guidance: %#v", got.transcript)
+	}
+}
+
+func TestBTWReturnRestartsParentSpinner(t *testing.T) {
+	for _, reducedMotion := range []bool{false, true} {
+		t.Run(map[bool]string{false: "animated", true: "reduced-motion"}[reducedMotion], func(t *testing.T) {
+			m := newBTWTestModel(t)
+			m.reducedMotion = reducedMotion
+			m.pending = true
+			m.spinnerTicking = true
+			side, _ := m.handleBTWCommand("")
+			side.spinnerTicking = false
+
+			returned, cmd := side.leaveBTW()
+			if !returned.pending || !returned.spinnerTicking {
+				t.Fatalf("parent spinner was not restarted: pending=%v ticking=%v", returned.pending, returned.spinnerTicking)
+			}
+			if cmd == nil {
+				t.Fatal("returning to an active parent did not schedule a spinner tick")
+			}
+		})
+	}
+}
+
+func TestBTWCancelledRunFlushesBeforeReturnAndRunIDsStayUnique(t *testing.T) {
+	m := newBTWTestModel(t)
+	m.provider = &fakeProvider{}
+
+	side1, _ := m.handleBTWCommand("first side question")
+	firstRunID := side1.activeRunID
+	side1.cancelRun()
+	blocked, _ := side1.leaveBTW()
+	if !blocked.btw.active || !transcriptContains(blocked.transcript, "still saving its session events") {
+		t.Fatal("BTW returned before the cancelled side run finished flushing")
+	}
+
+	flushedModel, _ := side1.updateModel(agentResponseMsg{runID: firstRunID})
+	flushed := flushedModel.(model)
+	if len(flushed.flushRunIDs) != 0 {
+		t.Fatalf("cancelled side run did not drain: %#v", flushed.flushRunIDs)
+	}
+	parent, _ := flushed.leaveBTW()
+	side2, _ := parent.handleBTWCommand("second side question")
+	if side2.activeRunID == firstRunID {
+		t.Fatalf("BTW run ID was reused across re-entry: %d", firstRunID)
+	}
+
+	updated, _ := side2.updateModel(agentResponseMsg{
+		runID: firstRunID,
+		rows:  []transcriptRow{{kind: rowAssistant, text: "answer to the FIRST question"}},
+	})
+	got := updated.(model)
+	if !got.pending || transcriptContains(got.transcript, "answer to the FIRST question") {
+		t.Fatal("a stale side response hijacked the next BTW run")
+	}
+}
+
+func TestBTWRejectsExplicitResumeOfSideSession(t *testing.T) {
+	m := newBTWTestModel(t)
+	side, _ := m.handleBTWCommand("")
+
+	if _, err := m.resolveResumeSession(side.activeSession.SessionID); err == nil || !strings.Contains(err.Error(), "not resumable") {
+		t.Fatalf("explicit side-session resume error = %v, want non-resumable rejection", err)
+	}
+}
+
+func TestBTWCannotStartDuringDeferredExit(t *testing.T) {
+	m := newBTWTestModel(t)
+	m.exiting = true
+
+	got, cmd := m.handleBTWCommand("should not run")
+	if cmd != nil || got.btw.active {
+		t.Fatalf("BTW started during deferred exit: active=%v cmd=%v", got.btw.active, cmd)
+	}
+	if !transcriptContains(got.transcript, "cannot start now") {
+		t.Fatalf("missing deferred-exit guidance: %#v", got.transcript)
+	}
+}
+
+func TestBTWStatusAndHelpStayVisible(t *testing.T) {
+	m := newBTWTestModel(t)
+	side, _ := m.handleBTWCommand("")
+
+	if status := plainRender(t, side.statusLine(100)); !strings.Contains(status, "BTW") {
+		t.Fatalf("status line has no persistent BTW indicator: %q", status)
+	}
+	groups := side.buildKeybindingGroups()
+	if got := groups[0].bindings[3].desc; !strings.Contains(got, "return to the main session") {
+		t.Fatalf("BTW Ctrl+C help = %q", got)
+	}
+}
+
+func TestBTWResizeUpdatesHiddenParent(t *testing.T) {
+	m := newBTWTestModel(t)
+	m.width = 80
+	m.height = 24
+	side, _ := m.handleBTWCommand("")
+
+	updated, _ := side.updateModel(tea.WindowSizeMsg{Width: 120, Height: 50})
+	got := updated.(model)
+	if got.btw.parent == nil || got.btw.parent.width != 120 || got.btw.parent.height != 50 {
+		t.Fatalf("hidden parent kept stale geometry: %#v", got.btw.parent)
+	}
+	returned, _ := got.leaveBTW()
+	if returned.width != 120 || returned.height != 50 {
+		t.Fatalf("restored parent geometry = %dx%d, want 120x50", returned.width, returned.height)
+	}
+}
+
+func TestBTWUsesIndependentUsageTracker(t *testing.T) {
+	m := newBTWTestModel(t)
+	parentTracker := m.usageTracker
+	side, _ := m.handleBTWCommand("")
+	if side.usageTracker == nil || side.usageTracker == parentTracker {
+		t.Fatal("BTW conversation shared the main session usage tracker")
+	}
+	returned, _ := side.leaveBTW()
+	if returned.usageTracker != parentTracker {
+		t.Fatal("returning from BTW did not restore the main session usage tracker")
+	}
+}
+
+func TestBTWFailedForkKeepsMainSessionLoops(t *testing.T) {
+	badRoot := filepath.Join(t.TempDir(), "sessions")
+	if err := os.WriteFile(badRoot, []byte("not a directory"), 0o600); err != nil {
+		t.Fatal(err)
+	}
+	m := newModel(context.Background(), Options{SessionStore: sessions.NewStore(sessions.StoreOptions{RootDir: badRoot})})
+	m.activeSession = sessions.Metadata{SessionID: "main-session"}
+	m.loops = []*loopState{{id: "loop-1"}}
+
+	got, _ := m.handleBTWCommand("")
+	if len(got.loops) != 1 || got.loops[0].id != "loop-1" {
+		t.Fatalf("failed BTW fork cleared main-session loops: %#v", got.loops)
+	}
+	if !transcriptContains(got.transcript, "Could not open BTW conversation") {
+		t.Fatalf("missing fork failure notice: %#v", got.transcript)
+	}
+}
+
+func TestBTWCommandReturnsToParentSession(t *testing.T) {
+	m := newBTWTestModel(t)
+	parentID := m.activeSession.SessionID
+	side, _ := m.handleBTWCommand("")
+
+	updated, cmd := side.dispatchCommand(parseCommand("/btw"))
+	returned := updated.(model)
+	if cmd != nil {
+		t.Fatal("returning from an idle BTW conversation should not start work")
+	}
+	if returned.btw.active || returned.btw.parent != nil {
+		t.Fatalf("BTW state remained active after /btw: %#v", returned.btw)
+	}
+	if returned.activeSession.SessionID != parentID {
+		t.Fatalf("returned session = %q, want parent %q", returned.activeSession.SessionID, parentID)
+	}
+}
+
+func TestBTWCtrlCDuringRunDoesNotClearDraft(t *testing.T) {
+	m := newBTWTestModel(t)
+	side, _ := m.handleBTWCommand("")
+	side.pending = true
+	side.input.SetValue("keep this draft")
+
+	updated, cmd := side.handleCtrlC()
+	got := updated.(model)
+	if cmd != nil {
+		t.Fatal("Ctrl+C should not start a command while a BTW response is running")
+	}
+	if !got.btw.active {
+		t.Fatal("Ctrl+C returned from BTW while its response was still running")
+	}
+	if got.composerValue() != "keep this draft" {
+		t.Fatalf("Ctrl+C cleared the in-flight BTW draft: %q", got.composerValue())
+	}
+	if !transcriptContains(got.transcript, "BTW response is still running") {
+		t.Fatalf("missing in-flight return guidance: %#v", got.transcript)
+	}
+}

internal/tui/command_center.go

@@ -214,6 +214,12 @@ func (m model) setDoctorStatusRow(text string) model {
 	for i := len(m.transcript) - 1; i >= 0; i-- {
 		if m.transcript[i].id == doctorStatusRowID {
 			m.transcript[i] = row
+			// Settled alt-screen rows are cached across frames; force a rebuild
+			// so an in-place update (e.g. the spinner tick) is reflected
+			// immediately instead of serving the stale cached snapshot.
+			if i < m.flushed {
+				m.altScreenSettledWidth = 0
+			}
 			return m
 		}
 	}
@@ -464,13 +470,15 @@ func (m model) handleModelCommand(args string) (model, string) {
 		// dedupe against the entry just persisted here, showing the same switch twice.
 		config.RecentModelEntry{Provider: m.providerName, Model: target.modelID},
 	)
-	resetEffort := false
-	if m.reasoningEffort != "" && !reasoningEffortAllowed(target.reasoningEfforts, m.reasoningEffort) {
-		// Drop an unsupported carry-over preference and fall back to the
-		// model's effective default for the new model.
-		m.reasoningEffort = ""
-		resetEffort = true
-	}
+	// Drop a known-unsupported preference, void any profile bookkeeping the
+	// drop erased, and re-derive an active profile's per-model effort fill.
+	// The ring is authoritative only for catalog-resolved targets
+	// (target.entry non-nil); live-discovered/custom targets carry no support
+	// knowledge, so an explicit preference survives onto them. resetEffort
+	// reports a dropped preference nothing refilled (shown as a reset to
+	// auto).
+	var resetEffort bool
+	m, resetEffort = m.reconcileEffortForModelSwitch(target.reasoningEfforts, target.entry != nil)
 	effortLine := "effort: " + m.effortDisplay()
 	if resetEffort {
 		// Preference was dropped: show "auto" (model default applies), not a
@@ -551,6 +559,13 @@ func (m model) switchProviderModel(providerName, modelID string) (model, string,
 	m.providerProfile = target
 	m.providerName = target.Name
 	m.modelName = target.Model
+	// An active profile's effort fill is per-model: re-derive it for the
+	// destination, exactly like handleModelCommand does. No generic
+	// unsupported-drop here: cross-provider targets are often custom models
+	// the catalog cannot vouch for either way, so an explicit preference is
+	// carried (pre-existing behavior) while the profile's own fill stays
+	// conservative — it only ever applies where support is known.
+	m = m.reconcileProfileAfterModelSwitch(m.availableReasoningEfforts())
 	// Record the outgoing pair too — see the matching comment in
 	// handleModelCommand for why (keeps the session's starting model from
 	// silently dropping out of "Recent" on the first switch away from it).
@@ -682,6 +697,14 @@ func (m model) persistSelectedModel(profile config.ProviderProfile) (bool, error
 	if model == "" {
 		return false, nil
 	}
+	persisted, err := config.ProviderPersisted(path, name)
+	if err != nil {
+		return false, err
+	}
+	if !persisted {
+		// Env-derived providers have no config.json row to update.
+		return false, nil
+	}
 	if _, err := config.SetProviderModel(path, name, model); err != nil {
 		return false, err
 	}

internal/tui/commands.go

@@ -41,11 +41,13 @@ const (
 	commandAddDir
 	commandSelfCorrect
 	commandTurns
+	commandProfile
 	commandRetry
 	commandEdit
 	commandCopy
 	commandExport
 	commandNew
+	commandBTW
 	commandSkills
 	commandLoop
 	commandVoice
@@ -191,6 +193,13 @@ var commandDefinitions = []commandDefinition{
 		description: "Start a fresh session; the current one stays resumable via /resume.",
 		kind:        commandNew,
 	},
+	{
+		name:        "/btw",
+		usage:       "/btw [question]",
+		group:       commandGroupSession,
+		description: "Open an isolated side conversation; run /btw again to return.",
+		kind:        commandBTW,
+	},
 	{
 		name:        "/search",
 		aliases:     []string{"/find"},
@@ -252,6 +261,7 @@ var commandDefinitions = []commandDefinition{
 	},
 	{
 		name:        "/rewind",
+		aliases:     []string{"/undo"},
 		usage:       "/rewind [latest|<sequence>]",
 		group:       commandGroupSession,
 		description: "Restore workspace files to a checkpoint and truncate the session.",
@@ -286,6 +296,13 @@ var commandDefinitions = []commandDefinition{
 		description: "Show or set the per-run tool-turn budget for this session (raise it for long multi-step tasks).",
 		kind:        commandTurns,
 	},
+	{
+		name:        "/profile",
+		usage:       "/profile [status|balanced|fast|thorough]",
+		group:       commandGroupSession,
+		description: "Show or switch the execution profile for the next run (loop posture: turn budget, effort, self-correction, escalation; model selection is unchanged).",
+		kind:        commandProfile,
+	},
 	{
 		name:        "/retry",
 		usage:       "/retry",

internal/tui/commands_test.go

@@ -985,6 +985,31 @@ func TestParseBackgroundTerminalCommands(t *testing.T) {
 	}
 }
 
+// /undo is an alias for /rewind (issue #698): previously it fell through to
+// commandUnknown.
+func TestUndoAliasesRewind(t *testing.T) {
+	rewind, ok := resolveCommand("/rewind")
+	if !ok {
+		t.Fatal("expected /rewind to resolve")
+	}
+	alias, ok := resolveCommand("/undo")
+	if !ok || alias.kind != rewind.kind {
+		t.Fatalf("expected /undo to alias /rewind, got ok=%v command=%#v", ok, alias)
+	}
+	// End-to-end: a typed /undo dispatches to the rewind handler with args intact,
+	// for both the "latest" and numeric-<sequence> forms of the /rewind contract.
+	if parsed := parseCommand("/undo latest"); parsed.kind != commandRewind || parsed.text != "latest" {
+		t.Fatalf("parseCommand(%q) = %#v, want commandRewind with text=latest", "/undo latest", parsed)
+	}
+	if parsed := parseCommand("/undo 123"); parsed.kind != commandRewind || parsed.text != "123" {
+		t.Fatalf("parseCommand(%q) = %#v, want commandRewind with text=123", "/undo 123", parsed)
+	}
+	// Guard: without the alias it would be commandUnknown.
+	if parsed := parseCommand("/undo"); parsed.kind == commandUnknown {
+		t.Fatal("/undo resolved to commandUnknown — the alias is not wired")
+	}
+}
+
 func TestCommandSelectionRequiresInputFromUsage(t *testing.T) {
 	cases := []struct {
 		name string

internal/tui/files_panel.go

@@ -348,12 +348,29 @@ func (m model) lastRowIndexForFile(path string) int {
 	return -1
 }
 
+func (m *model) setSelectedFile(path string) {
+	if m.selectedFile == path {
+		return
+	}
+	oldPath := m.selectedFile
+	m.selectedFile = path
+	for i := 0; i < m.flushed && i < len(m.transcript); i++ {
+		for _, changedPath := range m.transcript[i].changedFiles {
+			if changedPath == oldPath || changedPath == path {
+				m.altScreenSettledWidth = 0
+				return
+			}
+		}
+	}
+}
+
 // selectFile marks path as the selected file and scrolls the transcript so its
 // most recent edit card is in view; the card tint comes from the renderers
 // reading selectedFile (rowTouchesSelectedFile).
 func (m model) selectFile(path string) model {
-	m.selectedFile = path
-	if offset, ok := m.scrollOffsetForTranscriptRow(m.lastRowIndexForFile(path)); ok {
+	rowIndex := m.lastRowIndexForFile(path)
+	m.setSelectedFile(path)
+	if offset, ok := m.scrollOffsetForTranscriptRow(rowIndex); ok {
 		m.chatScrollOffset = offset
 		if offset == 0 {
 			m.chatBodyLines = 0

internal/tui/flush.go

@@ -69,13 +69,6 @@ func (m model) settledRow(row transcriptRow, rc rowContext) bool {
 // of the unflushed tail is rendered once and queued for scrollback. It returns
 // the (possibly nil) print command for the queued batch.
 func (m model) settleTranscript() (model, tea.Cmd) {
-	// In alt-screen mode there is no native scrollback surface for tea.Println:
-	// Bubble Tea drops that output by design. Keep rows in the managed view so
-	// the chat behaves like a fullscreen app and cannot reveal prior shell
-	// history by scrolling.
-	if m.altScreen {
-		return m, nil
-	}
 	// Never freeze history at the pre-WindowSizeMsg default width: the first
 	// real WindowSizeMsg arrives before any user-visible content settles, and
 	// flushing earlier would hard-wrap startup rows at 96 cols forever.
@@ -87,10 +80,14 @@ func (m model) settleTranscript() (model, tea.Cmd) {
 		// frontier themselves; this is a safety net).
 		m.flushed = len(m.transcript)
 	}
-	rc := buildRowContext(m.transcript)
+	oldFlushed := m.flushed
+	// An unsettled row blocks the frontier, so every cross-row dependency needed
+	// to decide or render the tail (call/result and prompt/decision pairs) also
+	// lives in the tail. Avoid rescanning the settled prefix on every tick.
+	rc := buildRowContext(m.transcript[m.flushed:])
 	width := chatWidth(m.width)
 	batch := []string{}
-	previousKind, havePreviousKind := previousVisibleTranscriptKind(m.transcript, m.flushed, rc)
+	previousKind, havePreviousKind := m.flushedPreviousKind, m.flushedHavePreviousKind
 	for m.flushed < len(m.transcript) {
 		row := m.transcript[m.flushed]
 		if !m.settledRow(row, rc) {
@@ -100,6 +97,15 @@ func (m model) settleTranscript() (model, tea.Cmd) {
 		if row.kind == rowWelcome || rc.skip(row) {
 			continue
 		}
+		// Bubble Tea intentionally discards tea.Println in alt-screen mode. The
+		// bookkeeping still advances; the settled prefix is retained in the body
+		// item cache rebuilt below.
+		if m.altScreen {
+			m.flushedAny = true
+			previousKind = row.kind
+			havePreviousKind = true
+			continue
+		}
 		rendered := m.renderRowMode(row, width, rc, true)
 		if rendered == "" {
 			continue
@@ -115,6 +121,15 @@ func (m model) settleTranscript() (model, tea.Cmd) {
 		previousKind = row.kind
 		havePreviousKind = true
 	}
+	m.flushedPreviousKind = previousKind
+	m.flushedHavePreviousKind = havePreviousKind
+	if m.altScreen {
+		bodyWidth := m.chatColumnWidth()
+		if m.flushed != oldFlushed || m.altScreenSettledWidth != bodyWidth || m.altScreenSettledFrontier != m.flushed {
+			m.rebuildAltScreenSettledItems(bodyWidth)
+		}
+		return m, nil
+	}
 	if len(batch) > 0 {
 		m.flushQueue = append(m.flushQueue, strings.Join(batch, "\n"))
 	}
@@ -139,6 +154,11 @@ func (m model) drainFlushQueue() (model, tea.Cmd) {
 // surface ended.
 func (m *model) resetFlushFrontier(divider string) {
 	m.flushed = 0
+	m.flushedPreviousKind = rowWelcome
+	m.flushedHavePreviousKind = false
+	m.altScreenSettledItems = nil
+	m.altScreenSettledWidth = 0
+	m.altScreenSettledFrontier = 0
 	// Renumbers every row's bodyY from the top (used by /clear, /resume, /rewind,
 	// /compact); a transcript-hover target's bodyY would otherwise risk
 	// coincidentally matching an unrelated row in the rebuilt transcript.
@@ -147,3 +167,35 @@ func (m *model) resetFlushFrontier(divider string) {
 		m.flushQueue = append(m.flushQueue, zeroTheme.faint.Render(divider))
 	}
 }
+
+// rebuildAltScreenSettledItems snapshots the stable prefix once when the
+// frontier (or width) changes. View then reuses these item descriptors and only
+// constructs descriptors/context for the live tail.
+func (m *model) rebuildAltScreenSettledItems(width int) {
+	if !m.altScreen || m.flushed == 0 {
+		m.altScreenSettledItems = nil
+		m.altScreenSettledWidth = width
+		m.altScreenSettledFrontier = m.flushed
+		return
+	}
+	snapshot := *m
+	snapshot.transcript = m.transcript[:m.flushed]
+	snapshot.flushed = 0
+	snapshot.flushedAny = false
+	snapshot.flushedHavePreviousKind = false
+	snapshot.altScreenSettledItems = nil
+	snapshot.altScreenSettledWidth = 0
+	snapshot.altScreenSettledFrontier = 0
+	snapshot.pending = false
+	snapshot.pendingSpecReview = nil
+	snapshot.fileView = fileViewState{}
+	built := snapshot.transcriptBodyItems(width, "", false)
+	// Keep scratch capacity for the live tail. transcriptBodyItems can then
+	// append transient rows without copying the entire settled prefix. The cache
+	// length never includes those scratch entries, and the next frame overwrites
+	// them on the single Bubble Tea render goroutine.
+	m.altScreenSettledItems = make([]transcriptBodyItem, len(built), len(built)+256)
+	copy(m.altScreenSettledItems, built)
+	m.altScreenSettledWidth = width
+	m.altScreenSettledFrontier = m.flushed
+}