ultimate-pi 0.16.0 → 0.18.0

package/.pi/harness/docs/adrs/0043-path-first-harness-tools.md

@@ -0,0 +1,38 @@
+# ADR 0043: Path-first harness tool contracts
+
+- **Status:** Accepted
+- **Date:** 2026-05-23
+
+## Context
+
+`approve_plan`, `create_plan`, and `submit_*` often pass full YAML/JSON documents in tool arguments when the same bytes already exist under `.pi/harness/runs/<run_id>/`. That duplicates tokens and trains models to carry large structs in chat history.
+
+## Decision
+
+1. **`approve_plan`** — Primary API: `approve_plan({ human_summary?: string })`. Extension loads `plan_packet` from `runCtx.plan_packet_path` and `research-brief.yaml` from the run dir. Optional `plan_packet` / `research_brief` deprecated for one release.
+2. **`create_plan`** — Primary API: `create_plan()` or `create_plan({ plan_packet_path?: string })`. Verifies approval marker and optional content hash from approve time.
+3. **`submit_*`** — Accept `source_path` under the active run; read, validate, promote to canonical path. `document` remains optional (deprecated).
+4. **`merge_harness_yaml`** — Parent merges patches from artifact paths without pasting bodies into tool args.
+5. **Tool results** — Return `{ path, sha256, status }` (and ids where relevant), not full documents.
+
+## Safety
+
+- Draft/canonical packet must exist on disk before approve.
+- Re-`approve_plan` required when `execution_plan` or `acceptance_checks` change after a `plan_gap` revise (hash gate).
+
+## Consequences
+
+### Positive
+
+- Approval turns stay small in session history.
+- Subagents write once to disk; submit is O(path) tokens.
+
+### Negative
+
+- Agents must write drafts before approve/submit (explicit discipline).
+
+## References
+
+- `.pi/extensions/harness-plan-approval.ts`
+- `.pi/extensions/lib/harness-subagent-submit-pipeline.ts`
+- ADR 0042, 0044

package/.pi/harness/docs/adrs/0044-harness-steer-loop.md

@@ -0,0 +1,36 @@
+# ADR 0044: Harness steer loop (post-run repair)
+
+- **Status:** Accepted
+- **Date:** 2026-05-23
+
+## Context
+
+After `/harness-run`, failed benchmarks or blocked execution previously routed users to `/harness-plan "<new task>"` even when the approved plan was still valid—high friction and duplicate planning context.
+
+## Decision
+
+1. **Always review** — `/harness-run` ends with `next_command: /harness-review` (including `blocked` / partial work). Remove benchmark fail-fast skip of verdict/adversary (ADR 0039 amended).
+2. **Review artifacts** — Parent writes `artifacts/review-outcome.yaml` and `artifacts/repair-brief.yaml` (path pointers, not pasted bodies).
+3. **Remediation routing** — `review-outcome.remediation_class`: `implementation_gap` → `/harness-steer`; `plan_gap` → `/harness-plan` revise with `repair_brief_path`; `pass` → policy status. **Review outcome wins** over executor `scope_drift` when they disagree; tie → `plan_gap`.
+4. **`/harness-steer`** — Thin orchestrator: read briefs, set policy **phase `execute`**, spawn `harness/executor` with `mode: repair`, then `/harness-review` again.
+5. **Caps** — `HARNESS_STEER_MAX_ATTEMPTS` (default 3). **Tiered review:** full review on initial run + steer 1; steers 2+ use lite (benchmark + verdict) unless prior `block_merge` or user forces full.
+6. **Sentrux** — Refresh baseline or compare new violations only after steer mutations (avoid false degraded on every attempt).
+7. **Evaluate-phase writes** — Orchestrator may write review/steer YAML under run `artifacts/` in `evaluate`/`adversary` phase (allowlisted files).
+
+## Consequences
+
+### Positive
+
+- One `approve_plan`; many repair cycles without re-typing tasks.
+- `harness-auto` can loop until pass or cap.
+
+### Negative
+
+- Higher review cost on failed runs (mitigated by tiered adversary).
+
+## References
+
+- `.pi/prompts/harness-steer.md`
+- `.pi/harness/specs/review-outcome.schema.json`, `repair-brief.schema.json`
+- `nextStepAfterOutcome` in `.pi/lib/harness-run-context.ts`
+- ADR 0039 (amended), 0043

package/.pi/harness/docs/adrs/README.md

@@ -24,6 +24,16 @@ Team-shared ADRs for the ultimate-pi harness live under `.pi/harness/docs/adrs/`
 | [0036](0036-implementation-research-and-selective-debate.md) | Implementation research and selective debate | Accepted |
 | [0037](0037-subagent-submit-tools.md) | Subagent submit tools (subprocess extension) | Accepted |
 | [0038](0038-budget-telemetry-only.md) | Budget caps telemetry-only by default | Accepted |
+| [0039](0039-harness-post-run-review-gate.md) | `/harness-review` master post-run gate | Accepted |
+| [0040](0040-practice-grounded-orchestration.md) | Practice-grounded orchestration & team topology | Accepted |
+| [0041](0041-intelligent-planning-reconnaissance.md) | Intelligent planning reconnaissance (tools over tool-scouts) | Accepted |
+| [0042](0042-agent-native-orchestration.md) | Agent-native orchestration (lakes, plan-verify probes, synthesizer) | Accepted |
+| [0043](0043-path-first-harness-tools.md) | Path-first harness tool contracts | Accepted |
+| [0044](0044-harness-steer-loop.md) | Post-run steer loop (repair vs plan revise) | Accepted |
+
+## Practice map
+
+Phase-to-practice mapping for slash commands: [practice-map.md](../practice-map.md).
 
 ## Template
 

package/.pi/harness/docs/graphify-kb-updater-runbook.md

@@ -0,0 +1,157 @@
+# Graphify KB updater runbook
+
+## Purpose and scope
+
+`graphify-kb-updater` keeps the local Graphify source corpus current for agentic engineering, context engineering, harness engineering, AI coding harnesses, research papers/feeds, articles/blogs, local books/transcripts, YouTube candidates, and competitor intelligence.
+
+The approved operating model is **hybrid allowlist auto-promotion with conservative staging**:
+
+- Daily local automation may auto-promote only explicitly approved allowlisted public sources with complete provenance and rights/access metadata.
+- Books, transcripts, YouTube/video material, paid/copyrighted/mirrored material, unclear-license content, and unknown open-web sources remain staged until manually approved.
+- Competitor monitoring is a curated taxonomy/watchlist/reporting signal, not an exhaustive crawler.
+- Pi-agent-open integration is intentionally limited/deferred: opening Pi should do at most a low-latency, no-network stale check. It must not perform synchronous web discovery, promotion, or Graphify mutation.
+
+## Governance and approval boundaries
+
+Required rights/access fields for every promotion:
+
+- `license`
+- `access`
+- `approved_by`
+- `approved_at`
+
+Allowlist auto-promotion requires all of the following:
+
+1. `.pi/harness/corpus/graphify-kb-updater.config.json` has `auto_promote_allowlist: true`.
+2. The candidate domain is present in `allowlist` with `approved: true`.
+3. The candidate itself has `approved: true`.
+4. `rights_access` is complete.
+5. The candidate is not a risky source class that requires manual review.
+
+Risky source classes (`book`, `transcript`, `youtube`) always require explicit approval and complete rights/access metadata. Raw HTTP shell paths are forbidden; keep discovery/fetch through approved harness web/API abstractions and verify with `.pi/scripts/harness-web-policy-guard.mjs`.
+
+## Manual commands
+
+Dry-run, no mutation of `raw/`, state, or `graphify-out/`:
+
+```bash
+node .pi/scripts/graphify-kb-updater.mjs --dry-run --pilot-report
+```
+
+Apply approved/promotable candidates and refresh Graphify only when promoted files changed:
+
+```bash
+node .pi/scripts/graphify-kb-updater.mjs --apply --refresh-graph --pilot-report
+```
+
+Apply without graph mutation:
+
+```bash
+node .pi/scripts/graphify-kb-updater.mjs --apply --skip-graph --pilot-report
+```
+
+Validate scheduler templates:
+
+```bash
+node .pi/scripts/graphify-kb-updater.mjs --scheduler-smoke
+```
+
+Run web-policy guard:
+
+```bash
+node .pi/scripts/harness-web-policy-guard.mjs
+```
+
+## Approval workflow
+
+1. Review dry-run JSON: candidate count, source counts, competitor labels, duplicate/skipped/blocked counts, stale warnings, planned promotions, and graph action.
+2. For a candidate, add it to `.pi/harness/corpus/graphify-kb-updater.config.json` `review_queue` with:
+   - `kind` (`article`, `paper`, `book`, `transcript`, or `youtube`)
+   - `title`
+   - `url` or `path`
+   - `approved: true`
+   - `rights_access` object with all required fields
+   - optional `competitor_labels` or provenance notes.
+3. For local files, you may place `<file>.rights.json` beside the source, but risky classes still require explicit approval before promotion.
+4. Run `--apply --refresh-graph`.
+5. Promoted sources land under `raw/graphify-kb-updates/<kind>/` with `.provenance.json` sidecars.
+
+## Daily scheduler setup
+
+Systemd user timer is the primary path and runs daily at 08:30 with randomized delay:
+
+```bash
+mkdir -p ~/.config/ultimate-pi ~/.config/systemd/user ~/.local/state/ultimate-pi
+cp .pi/harness/corpus/systemd/graphify-kb-updater.env.template ~/.config/ultimate-pi/graphify-kb-updater.env
+# edit UP_ROOT in the env file
+cp .pi/harness/corpus/systemd/graphify-kb-updater.service ~/.config/systemd/user/
+cp .pi/harness/corpus/systemd/graphify-kb-updater.timer ~/.config/systemd/user/
+systemctl --user daemon-reload
+systemctl --user enable --now graphify-kb-updater.timer
+systemctl --user list-timers graphify-kb-updater.timer
+```
+
+The service uses `flock`, `timeout 45m`, explicit env, append-only logs, and a non-overlap lock.
+
+Cron fallback is daily at 08:30; edit `UP_ROOT` and copy the line from `.pi/harness/corpus/cron.example` with `crontab -e`.
+
+## Reports, logs, and fields
+
+Apply runs write:
+
+- Registry: `.pi/harness/corpus/graphify-kb-updater-state/registry.json`
+- Per-run logs: `.pi/harness/corpus/graphify-kb-updater-state/logs/`
+- Scheduler logs: `~/.local/state/ultimate-pi/graphify-kb-updater.log` and `.err`
+
+Each run reports:
+
+- `last_run_at`
+- `candidate_count`, `promoted_count`, `blocked_count`, `skipped_count`, `duplicate_skips`, `failure_count`
+- `counts.by_kind`, `counts.by_source_type`, `counts.by_competitor_label`, `counts.allowlisted`
+- `stale_warnings`
+- `changed_existing_count` for same URL/path content changes
+- `graph.action`, `graph.exit_status`, and Graphify report path when refreshed
+- optional pilot metrics: `frontier_recall_proxy`, `promoted_precision_proxy`, `duplicate_noise_rate`, `graphify_success`
+
+Review these fields before enabling unattended mode and after every config change.
+
+## Troubleshooting
+
+- `missing_rights_access_approval`: add complete rights/access metadata.
+- `manual_approval_required`: set `approved: true` after source and rights review.
+- `duplicate_unchanged`: candidate was already promoted and content hash is unchanged.
+- `changed_existing_count > 0`: a stable URL/path changed content; review before relying on previous conclusions.
+- Graphify skipped: no promoted changes, `--skip-graph`, or no `--refresh-graph`.
+- Graphify failed: inspect `graph.stderr`, run `graphify update .` manually, and keep the scheduler disabled until fixed.
+- Scheduler did not run: check `systemctl --user status graphify-kb-updater.timer`, the env file path, and scheduler logs.
+- Overlap: lock path `%t/graphify-kb-updater.lock` or `/tmp/graphify-kb-updater.lock` prevents concurrent runs.
+
+## Disable
+
+```bash
+systemctl --user disable --now graphify-kb-updater.timer
+systemctl --user reset-failed graphify-kb-updater.service
+```
+
+Remove any cron line copied from `.pi/harness/corpus/cron.example`.
+
+## Rollback
+
+1. Disable systemd timer and remove cron line.
+2. Use registry/log promoted paths to remove or quarantine promoted files under `raw/graphify-kb-updates/`.
+3. Restore `.pi/harness/corpus/graphify-kb-updater-state/registry.json` from backup, or mark candidates rejected/quarantined.
+4. Revert implementation files if needed:
+
+```bash
+git checkout -- .pi/scripts/graphify-kb-updater.mjs .pi/harness/corpus/graphify-kb-updater.config.json .pi/harness/corpus/systemd/graphify-kb-updater.timer .pi/harness/corpus/cron.example test/graphify-kb-updater.test.mjs .pi/harness/docs/graphify-kb-updater-runbook.md
+```
+
+5. Regenerate Graphify from valid sources:
+
+```bash
+graphify update .
+```
+
+## Pilot gate before unattended mode
+
+Run at least one dry-run and one supervised apply. Record frontier recall proxy, promoted precision proxy, duplicate/noise rate, skipped reasons, stale warnings, and Graphify success from `--pilot-report`. Enable the timer only if promoted precision is acceptable and graph refresh succeeds.

package/.pi/harness/docs/practice-map.md

@@ -0,0 +1,110 @@
+# Harness practice map
+
+Source of truth linking harness phases to proven practices (graphify corpus), agents/scripts, spawn topology, and **agent translation** (ADR 0042). Orchestrators and agents should cite this doc when unsure why a lane exists.
+
+See also: [ADRs](adrs/README.md), [ADR 0040](adrs/0040-practice-grounded-orchestration.md), [ADR 0041](adrs/0041-intelligent-planning-reconnaissance.md), [ADR 0042](adrs/0042-agent-native-orchestration.md), [ADR 0043](adrs/0043-path-first-harness-tools.md), [ADR 0044](adrs/0044-harness-steer-loop.md), [`raw/modules/structured-planning.md`](../../../raw/modules/structured-planning.md).
+
+## Agent translation (human practice → agent design)
+
+| Human practice | Agent translation |
+|----------------|-------------------|
+| Meeting / chair | Parent as **scheduler + gate checker** only |
+| Fagan inspection rounds | **Schema-bound probes** + merge (`parallel_probes` profile) |
+| Two-pizza cap per batch | **Token/spawn budget** per phase (`harness-spawn-budget.ts`) |
+| RACI roles | **Disjoint prompt contexts**, not serial speakers |
+| WBS decomposition | **Lake-first `execution_plan`** (few outcomes, bundled context) |
+| Sprint / story points | **`executor_strategy` + lake `done_criteria`** |
+| Critical path | **`critical_path_lake_ids`** |
+| Replan on every failure | **Steer loop** (`implementation_gap`) vs **plan revise** (`plan_gap`) |
+| Tool payloads in chat | **Path-first** approve/submit/merge (ADR 0043) |
+
+## Team management rules (all `/harness-*` orchestrators)
+
+1. **Parallelism law** — Parallel `subagent` `tasks` only when outputs are independent inputs to a later merge (implementation ∥ stack research; inspector ∥ adversary in `parallel_probes`). Never parallelize decompose ∥ hypothesis.
+2. **Two-pizza cap per batch** — Max 2 research lanes, max 1 optional `planning-context` subagent, max 1 executor, max 1 debate lane agent per `subagent` call (plan-verify may use 2 probes + integrator in separate batches).
+3. **No redundant thinkers** — If artifact X exists, downstream agents read it; they do not re-derive (e.g. decompose after `planning-context.yaml`).
+4. **Sequential dependency chain** — planning context → problem framing / decompose → hypothesis → research → synthesis/author → DAG → plan-verify → approve → execute → review → (steer)* → policy.
+5. **Plan-verify (agent-native)** — For `fast`/`standard`, parallel probes then integrator; parent is chair, not participant. Threaded debate remains for `full` until parity.
+6. **Tool intelligence** — Parent chooses graphify, sg, ccc; subprocesses optional. **Path-first:** disk is source of truth; tool args are pointers (ADR 0043).
+
+## `/harness-plan` — Planning Process Group
+
+| Phase | Practice | Agent translation | Actor | Spawn |
+|-------|----------|---------------------|-------|-------|
+| 0 | Tooling / fast feedback | Pre-index once | Parent + `ccc` | Automatic |
+| 1 | Reconnaissance before WBS | **ContextPack** on disk | Parent tools or optional `planning-context` | No default subprocess |
+| 2a | Problem framing / lakes | Lake outcomes, not ticket tree | `decompose` or synthesizer section | Sequential after context gate |
+| 2b | Hypothesis-driven approach | Falsifiable claim grounded in framing | `hypothesis` or synthesizer | After `artifacts/decomposition.yaml` |
+| 3.5 | Spike / external research | Paths in research brief | Researchers optional | Artifacts required |
+| 4 | Fork resolution (batched) | One `ask_user` gate | Parent | After 3.5 |
+| 4b | Lake-first execution plan | `executor_strategy`, context bundles | `plan-synthesizer` (low/med) or `execution-plan-author` (high) | Single agent |
+| 4c | Deterministic quality gate | Script, not LLM | `validate-plan-dag.mjs` | Parent; hard stop |
+| 4d | Tailor process to risk | Probe depth, not meeting count | `harness_plan_debate_eligibility` | Pre plan-verify |
+| 4e | Architectural intent | Fitness-function spec | `harness/sentrux-steward` optional | When structural risk |
+| 5 | Plan-verify (Review Gate) | Parallel probes + integrator | Debate cast / probes | `parallel_probes` or threaded |
+| 6 | Baseline + approve | Path-only `approve_plan` | Parent | `approve_plan`, `create_plan` |
+
+### Review Gate — debate RACI (threaded / full profile)
+
+| Agent | Inspection role | Practice | When |
+|-------|-----------------|----------|------|
+| `hypothesis-validator` | Blind verifier | Independent verification (ADR 0034) | Round 1 / fast path |
+| `plan-evaluator` | Inspector | Neutral checklist | Every required focus |
+| `plan-adversary` | Red team | Adversarial review | Every required focus |
+| `sprint-contract-auditor` | DoD auditor | Sprint contract | `quality` focus |
+| `review-integrator` | Recorder | Single round artifact | End of round |
+| Parent | Chair | Gates only | Always |
+
+### Plan-verify profiles
+
+| Profile | When | Team shape |
+|---------|------|------------|
+| `full` | High risk, material fork | Threaded: all four focuses |
+| `standard` | Default med | `parallel_probes`: inspector ∥ adversary → integrator |
+| `light` | Low risk | Threaded: `spec` + `quality` |
+| `fast` | Med/low, clear stack | Consolidated verify + blind hypothesis-validator |
+
+## `/harness-run` — Executing Process Group
+
+| Step | Practice | Agent translation | Actor |
+|------|----------|-------------------|-------|
+| Gate | Change control | `plan_ready` required | Parent |
+| Pre-work | Fitness baseline | `sentrux gate --save` | Parent |
+| Work | Single implementer | `executor_strategy` | `harness/executor` |
+| Post-work | Observation | `sentrux check` / signal artifact | Parent |
+| Handoff | Generator–evaluator | `submit_executor_handoff` | Executor |
+| Next | Always verify | **`/harness-review`** (not replan on blocked) | Parent routing |
+
+## `/harness-review` — Monitoring and Controlling
+
+| Phase | Practice | Agent translation | Actor |
+|-------|----------|-------------------|-------|
+| 1 | Automated QC + fitness | Deterministic first | Parent scripts |
+| 2 | Measure vs plan | Benchmark on disk | `evaluator` benchmark |
+| 3 | Policy audit | Verdict (no fail-fast skip) | `evaluator` verdict |
+| 4 | Red team | Tiered: full attempt 1, lite 2+ steer | `adversary` |
+| 5 | Outcome + repair brief | Machine routing | Parent + `review-outcome.yaml`, `repair-brief.yaml` |
+| 6 | Steer gate | One `ask_user` | harness-decisions |
+| 7 | Steer / revise | `implementation_gap` → `/harness-steer`; `plan_gap` → plan revise | ADR 0044 |
+
+`--quick` = deterministic + benchmark + verdict (no adversary). Steer attempts 2+ default to lite review unless `block_merge`.
+
+## `/harness-steer` — Repair sub-cycle (ADR 0044)
+
+| Step | Practice | Actor |
+|------|----------|-------|
+| 0 | Read review + repair briefs | Parent |
+| 1 | Policy phase → `execute` | Parent |
+| 2 | Repair scope | `harness/executor` `mode: repair` |
+| 3 | Re-verify | `/harness-review` |
+
+## Anti-patterns
+
+- **Do not** spawn `decompose` and `hypothesis` in the same parallel `tasks` batch.
+- **Do not** run `graphify query` in `decompose` when planning-context coverage is ok (ADR 0041).
+- **Do not** parallelize threaded debate lanes in one batch (except `parallel_probes` inspector ∥ adversary per ADR 0042).
+- **Do not** let executor or parent self-certify.
+- **Do not** stop review on benchmark fail — complete verdict and route via steer (ADR 0044).
+- **Do not** tell user to run `/harness-plan "<new task>"` on test failure — use `/harness-steer` with `repair-brief.yaml`.
+- **Do not** re-`approve_plan` every steer attempt — only when packet changes.
+- **Do not** embed full plan packets in `approve_plan` / `submit_*` tool args (ADR 0043).

package/.pi/harness/env.harness.template

@@ -21,6 +21,8 @@ HARNESS_WEB_SEARCH_ENGINE=ddg_html
 
 # --- PostHog (optional) ---
 # Project key — required for harness_* telemetry when HARNESS_TELEMETRY_ENABLED=true
+# WSL2: ultimate-pi loads 00-posthog-network-bootstrap.ts (IPv4 fetch for *.posthog.com).
+# If flush still fails, set POSTHOG_ENABLED=false or fix outbound HTTPS to PostHog.
 # POSTHOG_API_KEY=
 # POSTHOG_HOST=https://us.i.posthog.com
 # POSTHOG_ENABLED=true
@@ -39,6 +41,6 @@ HARNESS_WEB_SEARCH_ENGINE=ddg_html
 # --- Wiki / Obsidian vault (optional) ---
 VAULT_WIKI_PATH=vault/wiki
 
-# --- Sentrux gate (optional) ---
-# Require Sentrux stub for harness-verify (see .pi/scripts/harness-verify.mjs)
-# HARNESS_SENTRUX_REQUIRED=true
+# --- Sentrux fitness functions ---
+# Require sentrux check + run signal (or CI stub) in harness-verify
+HARNESS_SENTRUX_REQUIRED=true

package/.pi/harness/evals/smoke/fixtures/plan-phase/minimal-med-fast/artifacts/implementation-research.yaml

@@ -0,0 +1,28 @@
+schema_version: "1.0.0"
+problem_framing: Validate harness plan-phase with fixture-driven smoke
+sub_problems:
+  - DAG validation
+  - Debate gate coverage
+internal_references:
+  - path: .pi/harness/evals/smoke/smoke-harness-plan.mjs
+    relevance: Existing smoke pattern
+    reuse_signal: high
+external_references: []
+solution_patterns:
+  - name: fixture-driven gate
+    provenance: in-repo smoke
+    fit: Validates plan pipeline without live agents
+    tradeoffs:
+      pros: [Deterministic CI]
+      cons: []
+    risks: []
+similar_implementations: []
+recommended_approach:
+  summary: Extend minimal-med fixture with implementation artifact
+  recommended_approach_confidence: high
+  confidence_rationale: Reuses established smoke-harness-plan pattern
+  evidence_refs:
+    - .pi/harness/evals/smoke/smoke-harness-plan.mjs
+    - .pi/scripts/validate-plan-dag.mjs
+anti_patterns: []
+open_questions: []

package/.pi/harness/evals/smoke/fixtures/plan-phase/minimal-med-fast/artifacts/review-round-consolidated.yaml

@@ -0,0 +1,25 @@
+schema_version: "1.0.0"
+round_index: 1
+debate_round_focus: all
+round_summary: Consolidated review gate for fast profile fixture
+validation_summary: Spec and quality checks pass in one round
+adversary_summary: No blockers
+disputes: []
+recommended_packet_patches: []
+review_gate_ready: true
+participants:
+  - PlanEvaluatorAgent
+  - PlanAdversaryAgent
+  - SprintContractAuditorAgent
+  - ReviewIntegratorAgent
+claims:
+  - consolidated review gate ready
+rebuttals: []
+evidence_refs: []
+token_usage:
+  per_agent:
+    PlanEvaluatorAgent: 120
+    PlanAdversaryAgent: 100
+    SprintContractAuditorAgent: 80
+  round_total: 300
+consensus_delta: 0.1

package/.pi/harness/evals/smoke/fixtures/plan-phase/minimal-med-fast/plan-packet.yaml

@@ -0,0 +1,196 @@
+schema_version: "1.0.0"
+contract_version: "1.1.0"
+plan_id: plan-smoke-fixture-001
+task_id: task-smoke-001
+scope: Smoke fixture for plan-phase harness validation with execution_plan and debate artifacts.
+assumptions:
+  - Fixture only; no live agent run
+risk_level: med
+acceptance_checks:
+  - id: AC-1
+    description: DAG validation passes
+  - id: AC-2
+    description: Consolidated debate round recorded (fast profile)
+  - id: AC-3
+    description: Stack brief present in research-brief
+  - id: AC-4
+    description: Sprint contract complete
+  - id: AC-5
+    description: plan-review.md renders
+rollback_plan:
+  revert_commit_ready: true
+  rollback_artifacts:
+    revert_command: git revert HEAD
+    revert_branch: main
+    patch_bundle: .pi/harness/runs/smoke-fixture/patch.bundle
+execution_plan:
+  schema_version: "1.0.0"
+  phases:
+    - phase_id: P1
+      name: Foundation
+      objective: Establish baseline and verify harness wiring
+      entry_criteria:
+        - Fixture loaded
+      exit_criteria:
+        - AC-1 satisfied
+      milestone: M1-baseline
+      work_item_ids: [WI-1, WI-2, WI-3]
+    - phase_id: P2
+      name: Build
+      objective: Implement core changes
+      entry_criteria:
+        - M1-baseline complete
+      exit_criteria:
+        - AC-2 satisfied
+      milestone: M2-build
+      work_item_ids: [WI-4, WI-5, WI-6]
+    - phase_id: P3
+      name: Verify
+      objective: Quality gate and documentation
+      entry_criteria:
+        - M2-build complete
+      exit_criteria:
+        - AC-5 satisfied
+      milestone: M3-ship
+      work_item_ids: [WI-7, WI-8]
+  work_items:
+    - work_item_id: WI-1
+      phase_id: P1
+      title: Load fixture packet
+      description: Read plan-packet.yaml from fixture directory
+      depends_on: []
+      files:
+        - .pi/harness/evals/smoke/fixtures/plan-phase/minimal-med/plan-packet.yaml
+      parallel_safe: true
+      done_criteria:
+        type: manual
+        spec: Fixture packet readable
+      acceptance_check_ids: [AC-1]
+    - work_item_id: WI-2
+      phase_id: P1
+      title: Run DAG validator
+      description: Execute validate-plan-dag.mjs
+      depends_on: [WI-1]
+      files:
+        - .pi/scripts/validate-plan-dag.mjs
+      parallel_safe: false
+      done_criteria:
+        type: command
+        spec: node .pi/scripts/validate-plan-dag.mjs --packet plan-packet.yaml
+      acceptance_check_ids: [AC-1]
+    - work_item_id: WI-3
+      phase_id: P1
+      title: Lint harness-yaml
+      description: Ensure YAML helpers parse fixture
+      depends_on: [WI-1]
+      files:
+        - .pi/lib/harness-yaml.ts
+      parallel_safe: true
+      done_criteria:
+        type: lint
+        spec: npm test
+      acceptance_check_ids: [AC-1]
+    - work_item_id: WI-4
+      phase_id: P2
+      title: Debate round 1-2 artifacts
+      description: Validate review-round YAML
+      depends_on: [WI-2]
+      files:
+        - .pi/harness/evals/smoke/fixtures/plan-phase/minimal-med/artifacts/review-round-r1.yaml
+      parallel_safe: false
+      done_criteria:
+        type: artifact
+        spec: artifacts/review-round-r1.yaml exists
+      acceptance_check_ids: [AC-2]
+    - work_item_id: WI-5
+      phase_id: P2
+      title: Debate round 3-4 artifacts
+      description: Validate final review round
+      depends_on: [WI-4]
+      files:
+        - .pi/harness/evals/smoke/fixtures/plan-phase/minimal-med/artifacts/review-round-r4.yaml
+      parallel_safe: false
+      done_criteria:
+        type: artifact
+        spec: artifacts/review-round-r4.yaml exists
+      acceptance_check_ids: [AC-2]
+    - work_item_id: WI-6
+      phase_id: P2
+      title: Stack research merge
+      description: research-brief includes stack section
+      depends_on: [WI-2]
+      files: []
+      non_code: true
+      parallel_safe: true
+      done_criteria:
+        type: manual
+        spec: research-brief.yaml contains stack key
+      acceptance_check_ids: [AC-3]
+    - work_item_id: WI-7
+      phase_id: P3
+      title: Sprint contract audit
+      description: R4 sprint audit artifact
+      depends_on: [WI-5]
+      files:
+        - .pi/harness/evals/smoke/fixtures/plan-phase/minimal-med/artifacts/sprint-audit-r4.yaml
+      parallel_safe: false
+      done_criteria:
+        type: artifact
+        spec: sprint-audit-r4.yaml present
+      acceptance_check_ids: [AC-4]
+    - work_item_id: WI-8
+      phase_id: P3
+      title: Render plan-review
+      description: Human-readable plan review markdown
+      depends_on: [WI-7]
+      files:
+        - .pi/harness/evals/smoke/fixtures/plan-phase/minimal-med/plan-review.md
+      parallel_safe: false
+      done_criteria:
+        type: manual
+        spec: plan-review.md non-empty
+      acceptance_check_ids: [AC-5]
+  sprint_contract:
+    in_scope:
+      - Fixture validation only
+    out_of_scope:
+      - Production deploy
+    definition_of_done: All smoke checks green
+    assumptions:
+      - CI environment has node
+    external_dependencies: []
+  wbs_dictionary:
+    - work_item_id: WI-1
+      deliverable: Fixture packet loaded
+      owner_role: executor
+      inputs: []
+      outputs: [parsed packet]
+  risk_register:
+    - risk_id: R1
+      description: DAG validator false negative
+      likelihood: low
+      impact: high
+      mitigation: Unit tests on validate-plan-dag.mjs
+      linked_work_item_ids: [WI-2]
+    - risk_id: R2
+      description: Debate cap misconfiguration
+      likelihood: med
+      impact: med
+      mitigation: debate-orchestrator plan profile tests
+      linked_work_item_ids: [WI-4]
+    - risk_id: R3
+      description: YAML parse drift
+      likelihood: low
+      impact: med
+      mitigation: harness-yaml strict parse
+      linked_work_item_ids: [WI-3]
+  schedule_metadata:
+    critical_path_work_item_ids: [WI-1, WI-2, WI-4, WI-5, WI-7, WI-8]
+    parallel_groups:
+      - [WI-1, WI-3]
+    schedule_baseline_note: Fixture topological order; no calendar dates
+  dag_validation:
+    status: pass
+    topological_order: [WI-1, WI-2, WI-3, WI-4, WI-5, WI-6, WI-7, WI-8]
+    cycles: []
+    conflicts: []

package/.pi/harness/evals/smoke/fixtures/plan-phase/minimal-med-fast/plan-review.md

@@ -0,0 +1,14 @@
+# Plan review (fixture)
+
+plan_id: plan-smoke-fixture-001
+
+## Execution plan
+
+Phases: P1 Foundation → P2 Build → P3 Verify
+
+Critical path: WI-1 → WI-2 → WI-4 → WI-5 → WI-7 → WI-8
+
+## Debate
+
+- Round 1 (spec): review_gate_ready
+- Round 4 (quality): review_gate_ready