pi-crew 0.9.62 → 0.9.65

package/CHANGELOG.md

@@ -2,6 +2,111 @@
 
 > **Note:** `atomic-write-v2.ts` / `AtomicWriter` mentioned in historical entries below was consolidated into `atomic-write.ts` as of v0.9.42. This changelog is preserved as historical record — the migration was completed (the v2 class was never adopted; v1 won on simplicity + symlink-safety + link+unlink atomicity). See `docs/migration/atomic-write-v2-migration.md` for the decision rationale.
 
+## [0.9.65] — team-tool schema empty-string guard (budgetTotal) + effectiveness empty-result guard + skill drift fix (2026-08-10)
+
+### Fixes
+- **`budgetTotal` empty-string unset marker accepted** (`src/schema/team-tool-schema.ts`): `budgetTotal` was the only numeric `TeamToolParams` field missing the `Literal("")` union branch that its siblings (`budgetWarning`, `budgetAbort`, `tokenBudget`, `interval`, `replyDeadline`) already had. Calling models that emit every schema key with defaults (documented behavior in `normalizeTeamParams`) were rejected by pi-ai's pre-handler `validateToolArguments` → `Validation failed for tool "team"` on every action. The `MISCONFIGURATION GUARD` (rejects 1-999) is preserved — only the unset marker is added. Caught by the Tier 9 feature battery in the `real-test-pi-crew` skill (Tiers 1-8 stayed green while the team tool was broken for emitting models).
+- **Effectiveness empty-result guard** (`src/runtime/effectiveness.ts`): a completed task with an EMPTY result artifact (`resultArtifact.sizeBytes === 0`) is now treated as no-observed-work, closing the monitoring gap where a child worker absorbed by a 429 rate-limit or model-not-found failure still emitted transcript/usage events and the run completed with `consistency=1` (zero real work). Empty-result tasks flow through the existing `noObservedWork` escalation (warn → blocked for mutating roles). New regression tests: `test/unit/runtime/core/effectiveness-guard.test.ts` (8 tests).
+- **`real-test-pi-crew` skill drift fix**: `PI_CREW_BROKER_DIAG_UI` / `src/ui/run-dashboard.ts:831` citation removed from `skills/real-test-pi-crew/SKILL.md` Tier 6 (the env var was removed in `e3ee6fe2`); Tier 6 now documents the screen-change-evidence replacement. `scripts/pty_probe.py` no longer sets the dead env var.
+
+### Verified
+- `npm run test:critical`: 101/101 pass (default, `PI_CREW_BROKER=0`, `PI_CREW_BROKER=1`).
+- New `effectiveness-guard.test.ts`: 8/8 pass.
+- `npm run typecheck` + `npm run build:bundle` exit 0; bundle md5 `e39373498d618de7e233c361ebb03b03`.
+- Full 9-tier real-test re-run (2026-08-10): Tiers 1-8 + 9a (10/10) + 9b (5/5 spawn paths) pass; the previous run's failure modes (model-not-found spawn loop, 429-absorbed empty output) did not reproduce. Report: `docs/real-test/reports/real-test-2026-08-10-full-9-tier-f4-effectiveness-guard.md`.
+
+## [0.9.64] — pi-rlm→pi-crew pattern transfer: worker scratchpad + crash-resume + cancellation + quick wins (2026-08-09)
+
+### Quick Wins (patterns 17/19/20/11 + spike CI)
+
+### Features
+- **Schema-driven docs (QW17)**: `agents/*.md` frontmatter `tools:` now matches the
+  enforced `ROLE_TOOL_CONFIGS` (4 drifts fixed); a sync-test pins the derivation;
+  `scripts/gen-role-tools-docs.mjs` renders `docs/role-tools.md` from the source.
+- **Retry-resume contract suite (QW19)**: pins `executeWithRetry` (`?`-glob, empty-
+  retryableErrors, maxAttempts:0, abort-during-sleep, default attemptId),
+  `FileCheckpointStore` (corrupt-file quarantine, list-skip, wrong-runId delete),
+  and exports `shouldRecoverTask` for direct testing.
+- **Failure-mode inventory (QW20)**: `docs/failure-mode-inventory.md` maps the 7
+  pi-rlm failure modes to pi-crew handlers (wedge gap closed by Phase 1
+  ping-before-execute; EPIPE/timeout interplay declared as gaps).
+- **Error-as-data contract (QW11)**: `evidenceStatusFor` + `attemptErrorFor`
+  extracted as pure functions from `runChildProcessTask`; contract test pins the
+  precedence (cancelled > failed > completed; E007 timedOut override; 429 gated).
+- **`test:spike` script**: wires the scratchpad spike tests into CI.
+
+
+### Phase 3 cancellation hardening (experimental)
+
+### Features
+
+- **Atomic snapshot write (D2')**: `engine.snapshotState` now writes via
+  `temp + rename` (same directory), eliminating the theoretical torn-write race
+  between the debounce timer and the F3 quit flush.
+- **Kill-and-restore verified**: the existing SIGTERM → pi print-mode dispose →
+  `session_shutdown` reason:"quit" → F3 flush chain already captures a fresh
+  snapshot on worker kill (no new handler required). Documented + pinned by a
+  gated real-worker test.
+- **EngineBusyError: deliberately skipped** — ping-before-execute already
+  detects a wedged guest; the FIFO execute queue makes a busy-reject redundant.
+
+
+### Phase 2 crash-resume (experimental)
+
+### Features
+
+- **Cross-attempt restore (crash-resume)**: a scratchpad worker attempt N+1 (retry
+  / crash-recovery re-queue / manual re-run) automatically revives the namespace
+  from the previous attempt's redacted snapshot artifact — via a single
+  spawn-time lookup (`findLatestScratchpadSnapshot`), no retry-loop / recovery
+  changes needed. Restores lazily on the first `execute` call (D7 lazy invariant
+  kept) with a one-line model-visible notice.
+  - Lookup: latest mtime (model-fallback index resets each retry round → number
+    is not write-order); tie-break lowest attempt = newest round.
+  - Security: read-time re-validation (containment + filename pattern + lstat +
+    size + mtime hint, D10); fail-open (D11) never breaks the worker; strict scan
+    (D12); base64 round-trip (D13); redacted secret → literal `"***"` (D4).
+- **Guest zombie backstop (SEC-1)**: the engine now overrides
+  `PI_CREW_PARENT_PID=worker pid` for the guest (D5), so an orphaned guest
+  (worker SIGKILL'd) is flagged by the zombie scanner — pure inheritance left
+  guests LIVE forever holding provider keys + broker token.
+- **Snapshot cap (D6)**: 4 MiB two-sided (write-side raw byteLength trim, read-
+  side file + per-var 256 KiB) bounds v8.deserialize amplification.
+
+
+### Phase 1 worker scratchpad (experimental)
+
+### Features
+
+- **Worker stateful scratchpad (experimental)**: opt-in persistent Bun-free JS
+  evaluator (`execute` tool) for workers, ported from the pi-rlm pattern.
+  State (variables, parsed data) compounds across `execute` calls within a task
+  attempt — intermediate results live in a namespace instead of being re-derived
+  from transcript text. Snapshot per-attempt into the run artifact store
+  (redacted + atomic) prepares the ground for Phase 2 crash-resume.
+  - Opt-in per role: `executor`, `test-engineer`, `verifier` (default on); other
+    roles via agent frontmatter `scratchpad: true` (write roles only).
+  - Security: S-6 read-only roles are gated out regardless of frontmatter
+    (privilege-elevation guard); F6 `scratchpad: false` is an explicit kill-switch;
+    raw snapshots never land in the artifact root (temp dir → redacted
+    `writeArtifact`).
+  - Dormant by default: only active when `PI_CREW_SCRATCHPAD=1` is set by the
+    spawner; zero behavior change for non-opt-in workers.
+
+
+## [0.9.63] — built-in performance observability + local-path provider-extension discovery (2026-08-08)
+
+### Features
+
+- **Built-in performance observability (byte-built-in, always-on).** Every team run now auto-attaches a detached resource sampler and auto-generates a detailed performance report on completion — no separate benchmark harness needed. Toggle per-team via frontmatter `observability: true|false` (default `true` for parsed team files; direct-object `TeamConfig` fixtures stay unset for test isolation).
+  - **Live resource sampler** (`scripts/resource-sampler.mjs`): samples CPU/RSS per-PID every 2s via ppid-tree attribution (root runner + all child workers, including respawns), with PID-reuse guard (`/proc` starttime), first-sample CPU exclusion, and **6 live warning categories** — `high_cpu` (≥300%), `rss_jump` (+200MB/interval), `rss_high` (≥1GB), `rss_leak` (window-30 monotonic +100MB), `proc_died`, `proc_zombie`. Rate-limited (10s/pid/category); `--no-live-warn` flag to silence.
+  - **Post-hoc analyzer** (`scripts/analyze-run.mjs`): combines `events.jsonl` + transcripts + `resources.jsonl` into a markdown report at `docs/perf-report-<runId>.md` with **22 anomaly categories** (task_failed, model_retry/model_cascade, large_gap, slow_phase, launch_delay, drain_stall, token_imbalance, no_cache, worker_respawn_churn, api_error_storm, zero_output_completion, sustained_cpu, transient_cpu/rss_spike, rss_growth, run_not_completed, high_failure_rate, run_idle, cost_unreported, missing_transcript, sampler_gap, tool_churn), a per-subagent timeline (launch / respawn / startup / active-work / drain / finalize), and token/cost/model attribution. Optional `--agents` flag emits per-agent breakdowns.
+  - **Runtime wiring** (`src/runtime/team-runner.ts`): `startPerfSampler` spawns the sampler detached + `unref`'d (death never affects the run); `schedulePerfAnalyze` runs the analyzer +3s after `after_run_complete` via an `unref`'d `setTimeout` (never blocks run completion). Strict `observability !== true` keeps test fixtures from spawning. The sampler auto-stops when the run manifest reaches a terminal status (`--watch-run` mode).
+  - **Overhead ≈ 0** (measured A/B): sampler ~0.05% of one core, 56MB RSS fixed; analyzer ~72ms one-shot after run; ~32KB artifacts/run. Verified end-to-end on real runs — see `docs/real-test/reports/real-test-2026-08-07-perf-obs-overhead.md`.
+
+### Bug fixes
+
+- **Local-path provider extensions were not discovered for child workers (oc-go and any `pi install <local-path>` provider went "Model not found").** `discoverProviderExtensions` only resolved `npm:` specs from `~/.pi/agent/settings.json` `packages`, skipping local-path specs (e.g. `../../source/my_pi/source/pi-other-provider`) on the assumption that local paths were the pi-crew extension itself. That assumption was wrong for local provider extensions: `--no-extensions` in `buildPiWorkerArgs` stripped the provider from every child spawn → every model from that provider hit `Error: Model "…" not found` → the fallback chain burned 5 spawn-fails (~10s) per task before landing on a builtin provider. Fix: resolve local-path specs (`./`, `../`, absolute) relative to the settings.json dir — same trust level as `npm:` (settings packages are a sanctioned channel written by `pi install`); skip the pi-crew package itself via `packageRoot()` (a worker must not re-load the orchestrator). **SEC-1 preserved** (project/project-pi AGENT extensions in `.crew/agents/*.md` frontmatter stay gated — that is a separate, untrusted channel). Tests in `test/unit/runtime/model/provider-extensions.test.ts`. Investigation + correction of the earlier "hidden models" mis-attribution in `docs/real-test/reports/real-test-2026-08-08-provider-ext-local-path.md`.
 
 ## [0.9.62] — provider-quota attribution per live-session agent + dead-worker alert re-fire fix (2026-08-06)
 

package/README.md

@@ -69,8 +69,10 @@ repo: https://github.com/baphuongna/pi-crew
 - **Durable event replay** (L1, v0.9.8) — `RunEventBus.onWithReplay()` catches up a re-subscribing dashboard/overlay with events it missed during transient absence (toggle, reconnect), replaying from the durable JSONL log with seq-based dedup. No information loss even if the live subscriber was briefly gone.
 - **Lossless-by-default output handling** (L4, v0.9.8) — worker output thresholds sized from measured data (100% of real outputs fit without compaction); when compaction is unavoidable it keeps head+tail (preserves closing code fences/headings) instead of head-only truncation. No more `[pi-crew compacted N chars]` markers eating the end of a worker's result.
 - **Inter-pi broker** (v0.9.47, default-on) — a Unix-domain-socket message bus that lets concurrently-running Pi sessions pass messages, steering notes, and task-status events to each other. **On by default** on Linux + macOS; auto-disabled on native Windows (no unix socket). Three independent kill switches: `broker.enabled: false` (config), `PI_CREW_BROKER=0` (env, always wins), Windows auto-disable. See [docs/decisions/2026-07-22-broker-phase4-gated-on.md](docs/decisions/2026-07-22-broker-phase4-gated-on.md).
+- **Worker stateful scratchpad** (experimental, opt-in per role) — `executor` / `test-engineer` / `verifier` workers get a `scratchpad` tool: a persistent Bun-free JS evaluator whose namespace **compounds across calls within a task attempt** (variables set in one cell are visible in the next), so intermediate results live in memory instead of being re-derived from the transcript. Snapshots are flushed (redacted, atomic) per-attempt into the artifact store; the next attempt (retry / crash-recovery re-queue / re-run) **automatically revives the namespace** from the latest snapshot. Dormant by default (armed only when the spawner sets `PI_CREW_SCRATCHPAD=1`); zero behavior change for non-opt-in workers. Ported from the `@shift-labs/pi-rlm` pattern. See [src/runtime/scratchpad/README.md](src/runtime/scratchpad/README.md) (Phase 1-3 design, env keys, guards, threat model).
 - **`test:critical` + `real-test-pi-crew` skill** (v0.9.47) — a curated 14-file / 97-test subset (`npm run test:critical`, ~20s) for fast in-loop verification, plus a bundled skill distilling the full 8-tier end-to-end verification discipline (unit → 3-path kill-switch proof → typecheck/bundle → live TUI probing → smoke team run). Prevents the verifier-worker hang that full `npm test` (>4 min) caused against the 300s worker timeout.
-- **Provider extensions in subagents** (v0.9.57) — pi-crew spawns child-pi workers with `--no-extensions` (security posture), which made extension-registered providers (e.g. `pi-commandcode-provider`) unresolvable inside subagents. pi-crew now **auto-discovers provider packages** from `~/.pi/agent/settings.json` `packages` (npm: specs) and loads them via `--extension` in every builtin/user subagent — so **all provider models work in subagents**. An explicit `runtime.agentExtensions: string[]` config is an optional extra allowlist on top of auto-discovery. **SEC-1 preserved:** project/project-pi agents never receive these (env-gate unchanged).
+- **Provider extensions in subagents** (v0.9.57, local-path support v0.9.63) — pi-crew spawns child-pi workers with `--no-extensions` (security posture), which made extension-registered providers (e.g. `pi-commandcode-provider`, `pi-other-provider`) unresolvable inside subagents. pi-crew now **auto-discovers provider packages** from `~/.pi/agent/settings.json` `packages` — both `npm:` specs (v0.9.57) and **local-path specs** like `../../source/foo` (v0.9.63) — and loads them via `--extension` in every builtin/user subagent, so **all provider models work in subagents**. An explicit `runtime.agentExtensions: string[]` config is an optional extra allowlist on top of auto-discovery. **SEC-1 preserved:** project/project-pi agents never receive these (env-gate unchanged).
+- **Built-in performance observability** (v0.9.63) — every team run auto-attaches a detached resource sampler (per-PID CPU/RSS via ppid-tree attribution, 6 live warning categories) and auto-generates a markdown performance report on completion (22 anomaly categories, per-subagent timeline, token/cost/model attribution). Toggle per-team via frontmatter `observability: true|false`. Overhead ≈ 0 (sampler ~0.05% CPU / 56MB RSS; analyzer ~72ms post-run). See [Built-in performance observability](#built-in-performance-observability) below.
 
 ---
 
@@ -287,6 +289,18 @@ The advisory is **informational only** — there is no `force:true` flag needed
 
 ## Recent changes
 
+### v0.9.65: team-tool schema empty-string guard + effectiveness empty-result guard (2026-08-10)
+
+- **`budgetTotal` empty-string unset marker accepted**: `budgetTotal` was the only numeric `TeamToolParams` field missing the `Literal("")` union branch its siblings had. Calling models that emit every schema key with defaults were rejected by pi-ai's pre-handler validation → `Validation failed for tool "team"` on every action. The `MISCONFIGURATION GUARD` (rejects 1-999) is preserved. Caught by the Tier 9 feature battery — Tiers 1-8 stayed green while the team tool was broken for emitting models.
+- **Effectiveness empty-result guard**: a completed task with an empty result artifact (`sizeBytes === 0`) is now treated as no-observed-work — closing the monitoring gap where a 429-absorbed / model-not-found worker produced zero real content but the run still completed with `consistency=1`. Empty-result tasks flow through the existing `noObservedWork` escalation. Regression tests: `test/unit/runtime/core/effectiveness-guard.test.ts` (8 tests).
+- Full 9-tier real-test re-run (2026-08-10): Tiers 1-8 + 9a (10/10) + 9b (5/5) pass; previous failure modes did not reproduce. See [CHANGELOG.md](CHANGELOG.md) §0.9.65 and `docs/real-test/reports/real-test-2026-08-10-full-9-tier-f4-effectiveness-guard.md`.
+
+### v0.9.63: built-in performance observability + local-path provider-extension discovery
+
+- **Built-in performance observability (always-on, toggle per team)**: every team run now auto-attaches a detached resource sampler (`scripts/resource-sampler.mjs` — per-PID CPU/RSS via ppid-tree attribution, 6 live warning categories: high_cpu / rss_jump / rss_high / rss_leak / proc_died / proc_zombie) and auto-generates a markdown performance report (`scripts/analyze-run.mjs` → `docs/perf-report-<runId>.md` — 22 anomaly categories, per-subagent launch/respawn/active/drain timeline, token/cost/model attribution). Runtime wiring in `src/runtime/team-runner.ts`: `startPerfSampler` (detached + `unref`'d; death never affects the run) + `schedulePerfAnalyze` (+3s after `after_run_complete`, `unref`'d `setTimeout`). Toggle: team frontmatter `observability: true|false` (default `true`). **Overhead ≈ 0** (A/B verified: sampler ~0.05% CPU / 56MB RSS, analyzer ~72ms post-run, ~32KB artifacts/run). New scripts: `scripts/resource-sampler.mjs`, `scripts/analyze-run.mjs`. Tests: `test/unit/scripts/{analyze-run,resource-sampler}-audit.test.ts`.
+- **Local-path provider extensions now discovered for child workers**: `discoverProviderExtensions` previously resolved only `npm:` specs from `~/.pi/agent/settings.json` `packages`, skipping local-path specs on the assumption they were the pi-crew extension itself. That broke local provider extensions (e.g. `pi-other-provider` installed via `pi install <local-path>`) — every model from such a provider hit `Error: Model "…" not found` in child workers and burned ~10s/task of fallback churn. Fix: resolve `./`, `../`, and absolute specs relative to the settings.json dir (same sanctioned trust level as `npm:`), and skip pi-crew itself via `packageRoot()`. **SEC-1 preserved** (project/project-pi AGENT extensions stay gated). Tests in `test/unit/runtime/model/provider-extensions.test.ts`.
+- See [CHANGELOG.md](CHANGELOG.md) §0.9.63 and `docs/real-test/reports/real-test-2026-08-08-provider-ext-local-path.md`.
+
 ### v0.9.57: team-tool schema repair + provider-extension auto-discovery + post-reorg repo-layout consolidation
 
 - **Team tool repaired (was broken live while tests stayed green)**: calling models emit empty-string/boolean defaults for every schema key, which pi-ai's pre-handler `validateToolArguments` rejected (`Validation failed for tool team`) and `Type.Unsafe` schema fields without `[TypeBox.Kind]` made `Value.Check` throw (`Unknown type`). Schema now accepts unset markers natively; `SkillOverride`/`FreeformConfig` switched to TypeBox-native constructors; `normalizeTeamParams` drops empties in the handler. Chain-runner also fixed (quote-aware step splitting). Caught by a new **Tier 9 feature battery** in the [`real-test-pi-crew`](skills/real-test-pi-crew/SKILL.md) skill (live team-tool action coverage).
@@ -363,6 +377,38 @@ pi-crew supports multiple runtime modes for task execution:
 { "executeWorkers": false }
 ```
 
+## Built-in performance observability
+
+Every team run auto-attaches a **detached resource sampler** and auto-generates a **performance report** on completion — measuring real resource usage and surfacing anomalies from the run's actual events/transcripts, not a synthetic benchmark.
+
+### Artifacts produced (per run, under `.crew/artifacts/<runId>/`)
+
+| File | Contents |
+|------|----------|
+| `resources.jsonl` | Per-PID CPU/RSS samples every 2s (root runner + all child workers via ppid-tree attribution, including respawns). PID-reuse guarded by `/proc` starttime; first-sample CPU excluded from averages. |
+| `perf-obs.log` | Sampler diagnostics: spawn marker, live warnings, terminal-stop confirmation. |
+| `docs/perf-report-<runId>.md` | Markdown report: 22 anomaly categories, per-subagent timeline (launch/respawn/startup/active-work/drain/finalize), token/cost/model attribution. |
+
+### Live warnings (written to `perf-obs.log` during the run)
+
+`high_cpu` (≥300% one core) · `rss_jump` (+200MB/interval) · `rss_high` (≥1GB) · `rss_leak` (window-30 monotonic +100MB) · `proc_died` · `proc_zombie`. Rate-limited (10s/pid/category).
+
+### Toggle
+
+```yaml
+# teams/my-team.team.md
+---
+name: my-team
+observability: false   # default: true — set false to skip sampler + report
+---
+```
+
+`observability: true` is the default for parsed team files; direct-object `TeamConfig` fixtures (unit tests) stay unset so they never spawn the sampler. `schedulePerfAnalyze` runs the analyzer `+3s` after `after_run_complete` via an `unref`'d `setTimeout` — it never blocks run completion. The sampler auto-stops when the run manifest reaches a terminal status.
+
+### Overhead
+
+Measured A/B (same team/goal, observability on vs off): **no detectable wall-time difference** (delta inside the 429-storm noise). Sampler: ~0.05% of one core, 56MB RSS fixed, detached + `unref`'d. Analyzer: ~72ms one-shot after run. ~32KB artifacts/run. See `docs/real-test/reports/real-test-2026-08-07-perf-obs-overhead.md`.
+
 ## Async Runs
 
 Async runs are **detached** from the session — they survive session switches and reloads. Pi-crew notifies when complete.

package/agents/critic.md

@@ -5,7 +5,7 @@ model: false
 systemPromptMode: replace
 inheritProjectContext: true
 inheritSkills: false
-tools: read, grep, find, ls
+tools: read, grep, find, ls, glob
 ---
 
 You are a critical reviewer. Find flaws, missing steps, unsafe assumptions, overengineering, underengineering, and verification gaps. Return concrete fixes to the plan.

package/agents/explorer.md

@@ -5,7 +5,7 @@ model: false
 systemPromptMode: replace
 inheritProjectContext: true
 inheritSkills: false
-tools: read, grep, find, ls
+tools: read, grep, find, ls, glob, bash
 ---
 
 You are a fast codebase explorer. Map relevant files, symbols, data flow, and constraints. Do not modify files. Return concise findings with paths and evidence.

package/agents/planner.md

@@ -5,7 +5,7 @@ model: false
 systemPromptMode: replace
 inheritProjectContext: true
 inheritSkills: false
-tools: read, grep, find, ls
+tools: read, grep, find, ls, glob
 ---
 
 You are a planning specialist. Convert the goal and discovery notes into a concrete, ordered plan. Identify dependencies, risks, validation steps, and handoff instructions for implementers.

package/agents/reviewer.md

@@ -5,7 +5,7 @@ model: false
 systemPromptMode: replace
 inheritProjectContext: true
 inheritSkills: false
-tools: read, grep, find, ls, bash
+tools: read, grep, find, ls, glob, bash
 ---
 
 You are a code reviewer. Review the implementation for bugs, regressions, maintainability issues, missing tests, and project-rule violations. Return prioritized findings with evidence.

package/agents/security-reviewer.md

@@ -5,7 +5,7 @@ model: false
 systemPromptMode: replace
 inheritProjectContext: true
 inheritSkills: false
-tools: read, grep, find, ls, bash
+tools: read, grep, find
 ---
 
 You are a security reviewer. Look for injection, authn/authz flaws, insecure defaults, secret exposure, unsafe filesystem/network behavior, and dependency risks. Return severity and remediation.

package/agents/test-engineer.md

@@ -5,7 +5,7 @@ model: false
 systemPromptMode: replace
 inheritProjectContext: true
 inheritSkills: false
-tools: read, grep, find, ls, bash, edit, write
+tools: read, edit, write, bash, ls
 ---
 
 You are a test engineer. Identify the right test level, add or adjust tests when asked, detect flaky assumptions, and report exact validation commands and results.

package/agents/writer.md

@@ -5,7 +5,7 @@ model: false
 systemPromptMode: replace
 inheritProjectContext: true
 inheritSkills: false
-tools: read, grep, find, ls, edit, write
+tools: read, edit, write, ls
 ---
 
 You are a documentation specialist. Produce clear, concise, maintainable docs and summaries. Preserve technical accuracy and avoid marketing fluff.