@azure-id/orc 2.0.2 → 2.1.1

package/templates/skills/_shared/gotchas.md

@@ -185,7 +185,19 @@ cited one), then ONE trace line:
 `FINDING-OUTCOME addressed=<n> disputed=<n> wontfix=<n> open=<n> pre=<n> suppressed=<n> :: <cat>:<addressed>/<total>,…`.
 Where: `/orc` and `/orc-ultra` at Phase 7 (`phases/summary.md`), `/orc-mini`
 after the risk option's review, `/orc-quick` after the review offer. A review
-with zero findings records nothing and is a HEALTHY answer.
+with zero findings records no observation and is a HEALTHY answer: its
+`FINDING-OUTCOME` line (or the hook `RETURN orc-reviewer-*` line) already counts
+it as a review in `orc gotcha quality` (v2.1.0). An `author: orc` observation
+without `run` is refused (exit 2) — a `ref` that starts with the run id fills it.
+
+**The reviewed ranges (v2.1.0, DE-13).** In the same step, store what the review
+READ: `orc gotcha observe -` with
+`{"kind": "review-scope", "run": "<trace name>", "commit": "<HEAD>", "files": {"<path>": [[start, end], …]}}`
+— the slice's `diff_ranges`, a clean review too. The CLI keeps the last 500
+reviews in `.claude/orc/review-scope.jsonl` (user data, never pruned by age) and
+counts what it drops. A later `orc fix record` on lines inside a stored range
+of an EARLIER run is a computed `miss`: the fix is promoted at 1 case and goes
+first in the next card.
 
 ### Per-lane mechanics (the spines keep the trigger; this is the detail)
 
@@ -205,16 +217,24 @@ file is never a reason to fall back, and gotchas here are purely additive: probe
 inject the scope-matching entries into the F2 slice, and stop there. It never
 writes one, because one executor plus one repair round is too little signal to
 attribute a cause — and a lane with no analyst writing repair memory is how that
-memory fills with guesses.
+memory fills with guesses. **A review is different (v2.1.0):** when `/orc-fast` runs
+a review (the F4 offer, or the user asks for one), it records that review's outcomes
+as "Review close" says — an observation is REVIEW LEARNING, not a repair-memory
+entry, and every lane that ran a review records it.
 
 **`/orc-quick` — reads and records, since 2.0.0 (a deliberate reversal).**
-Record at step 3.3, after entry N is appended, and only for a request with
-programmatic proof: a defect entry with `REPRO red` then `REPRO green`
+Record after the step 3.5 offers (v2.1.0 — it was 3.3, before the review offer
+existed to record), and only for a request with programmatic proof: a defect entry with `REPRO red` then `REPRO green`
 (`source: defect`, `reproduced: true`) · a PR thread the request took and fixed
 (`source: pr`) · a Sonar or SARIF item fixed (`source: sonar` / `sarif`) · a CI
-failure fixed (`source: ci`) · a review-offer P0/P1 fixed (`source: review`,
-`addressed`) · a review finding the user called "not a problem here"
-(`source: review`, `disputed`). A plain fix with none of these records nothing.
+failure fixed (`source: ci`) · every finding of a review offer, ONE observation per finding
+with its outcome, as "Review close" says (`source: review`; "not a problem here"
+= `disputed`). A plain fix with none of these records nothing.
+A FIX of a named finding (a Sonar issue, a failed check, a defect ORC or an AI
+tool caused) goes through `orc fix record -` (`via: "orc-quick"`, v2.1.0): the
+author is the finder, never `orc`; a Sonar issue key gives the importer's own id,
+so a later gotcha sync completes the same row; and the CLI computes `miss`
+from `review-scope.jsonl`.
 Why the reversal: quick is where most of the fixes happen, so a memory that
 skips it never learns; a PR thread is a human's judgement and a red → green
 repro is proof, so the v1 "too little signal" objection does not apply; and the

package/templates/skills/_shared/habits.md

@@ -74,14 +74,14 @@ compiled flow's last phase. `/orc-wait` makes no proposal.
   question; its `announce[]` line names the habit and its undo. A habit
   toward `no` is a pre-mark only, never applied. A learned
   `review_before_push: on` → the end-of-run batch ALWAYS carries option *a*
-  (dispatch `orc-reviewer-opus-5-med` on the diff), marked `→ usual`, even with
+  (dispatch `orc-reviewer-opus-5-low` on the diff), marked `→ usual`, even with
   no `risk: high` row. Record `any.ship.review-before-push` whenever option *a*
   is offered: `chose=review` when the user takes it, else `push`.
 - **`/orc-fast`.** `fast.f0.stale-wiki`: a habit NEVER suggests `continue`
   (a stale wiki is a risk the user takes each time). The CLI never proposes
   it (`never_option`); a suggestion row that names it anyway is ignored. Show
   a `refresh` or `mini` suggestion only, and never apply one. A learned `review_before_push: on` → F4
-  offers a dispatch of `orc-reviewer-opus-5-med` on the diff FIRST (P0/P1 block
+  offers a dispatch of `orc-reviewer-opus-5-low` on the diff FIRST (P0/P1 block
   the commit), marked `→ usual`; the reviewer is not in the fast pipeline, so
   it runs only on that yes. Record `any.ship.review-before-push` whenever it is
   offered.

package/templates/skills/_shared/opus5-only.md

@@ -1,137 +1,137 @@
-# Shared contract — Opus-5-only dispatch mode (hard-gated, forcing)
-
-Canonical rule for routing EVERY dispatched role to Opus 5. Consumed by the full
-orc spine, the ultra lane, and every lane with a fixed-role dispatch; the lint
-pins the `opus5-only.md` pointer into each of them.
-
-## The gate
-
-Read the resolved config at run start (`config.md` defaults ← `orc.config.yaml`).
-The mode is **hard-gated**: absolutely nothing changes unless `opus5_only: true`.
-When it is false (the default), dispatch every role exactly as today — this file
-is inert.
-
-## The mapping (only when `opus5_only: true`)
-
-**Scored executors** — one model, EFFORT as the cost dial, replacing the 6-band
-mixed-model table:
-
-| Score | Executor agent |
-|-----------|---------------------------|
-| `[0,90)`   | `orc-executor-opus-5-low`  |
-| `[90,100]` | `orc-executor-opus-5-med`  |
-
-Two bands, sharing the 90 edge with the default table's top two rows. Since
-v1.0.0 the default table's high end is ALREADY Opus 5.5 with effort as the dial,
-so this mode differs from it only below 65 — a third band here would be a
-distinction the default table stopped making.
-
-**Fixed roles** — dispatch the Opus 5.5 variant **instead of** the default role
-agent: same task slice, same return contract, same phase.
-
-| Role | Default agent | Opus-5-only variant |
-|------|---------------|---------------------|
-| mini executor | `orc-executor-sonnet-5-high` | `orc-executor-opus-5-low` |
-| fast executor | `orc-executor-sonnet-4-6-high` | `orc-executor-opus-5-low` |
-| mini analyze | `orc-analyze-mini-sonnet-5-high` | `orc-analyze-mini-opus-5-med` |
-| mini plan | `orc-planner-mini-sonnet-5-high` | `orc-planner-mini-opus-5-med` |
-| scout | `orc-scout-sonnet-4-6-high` | `orc-scout-opus-5-low` |
-| pattern codify | `orc-pattern-codifier-sonnet-5-high` | `orc-pattern-codifier-opus-5-med` |
-| wiki scan | `orc-wiki-scanner-opus-4-8-high` (deep) **or** `orc-wiki-scanner-sonnet-5-high` (light) | `orc-wiki-scanner-opus-5-med` |
-| claude write | `orc-claude-writer-opus-4-8-high` | `orc-claude-writer-opus-5-med` |
-| retro mine | `orc-retro-sonnet-5-high` | `orc-retro-opus-5-med` |
-
-The nine roles already pinned to `claude-opus-5-5` — analyst, planner, reviewer,
-verifier, test-author, combiner, learn-writer, advisor, judge — are already
-compliant and dispatch unchanged (their efforts are NOT rewritten by this mode).
-
-**The wiki scan tier ladder (v0.46.0) adds NO row here, and needs no new pair.**
-Off, the ladder picks deep or light per delta. ON, this mode already forces the
-wiki scanner to `orc-wiki-scanner-opus-5-med` — so **both tiers collapse onto that
-one shipped agent** and the ladder simply stops applying. A cheaper Opus 5.5 scanner
-variant for the light tier does not exist and must never be added: a pair for a
-tier that cannot occur while the flag is on is exactly the phantom this table
-exists to prevent. Ladder: `../orc-wiki/references/partial-refresh.md`.
-
-## Out of scope — never forced
-
-- **`orc-trace-writer-haiku-4-5` stays Haiku 4.5.** It transcribes a packet the
-  orchestrator hands it — no reasoning, no source reads. It is never in the
-  roster.
-- **`orc-graph-noter-sonnet-4-6-med` is not dispatched at all.** It has no Opus 5.5
-  variant: under the mode a code lane skips graph notes and prints
-  `graph notes: skipped (opus5_only)` (`code-graph.md` §6). The graph's structure
-  layer costs no model, so nothing else about the graph changes.
-- **`orc-diy`.** Its score table is compile-owned (`orc diy compile` →
-  `flow.lock.json`); a DIY flow dispatches whatever its lock says. Re-compile to
-  change it.
-- **`orc-quick` — the ONE exception to the flat precedence below.** Its dispatch
-  gate asks the USER which agent to spawn before every single dispatch, and that
-  hard gate is the lane's entire premise. A forcing mode that collapsed the menu
-  to one option would silently delete it, so the mode is fully INERT there:
-  `opus5_only` is neither read nor honored by `orc-quick`. When the mode is on,
-  the lane SAYS so at the gate (`orc-quick ignores opus5_only — both options are
-  live`), because a shadowed setting must never be silent in either direction.
-- **`/orc-challenge` — UNAFFECTED, not exempt.** All three of its agents are
-  already `claude-opus-5-5` (judge high, advisor medium, reader low), so the mode
-  has nothing to force: zero new pairs, no rename churn, no roster row. The
-  reader's `low` effort is a MEASUREMENT choice, not a cost one — a
-  harder-thinking cold reader reasons around exactly the gaps D4 exists to find
-  — so nothing may ever "upgrade" it.
-  See `orc-quick/references/dispatch-gate.md`.
-
-## It FORCES — precedence is flat while ON
-
-`opus5_only: true` outranks every other dispatch selector:
-
-- each role's default frontmatter pin;
-- **`rubric_bands_override`** — a hand-written executor table is ignored while
-  the mode is on.
-
-Turning the mode off restores both. `rubric_bands` remains granularity
-only, never a preset selector, in both modes.
-
-The forcing is flat across every lane that HONORS the mode — `orc-quick` does
-not, per "Out of scope" above.
-
-### `orc extra` sits ABOVE the mode, and it is not a fourth thing it overrides
-
-Extra decides **whether a Claude agent runs at all**. `opus5_only` and the score
-tables only decide WHICH Claude agent runs where extra did not take it. So the
-mode is neither shadowed nor shadowing — it is simply **NOT CONSULTED** for the
-work extra took:
-
-- **a scored task** — an extra route row covering that score outranks the mode;
-  every score no row covers still resolves on the Opus 5.5 ladder.
-- **a POSITION** (v0.55.0 — `quick-executor`, `fast-executor`, `doc-writer`,
-  `doc-checker`, `wiki-scanner-deep`, `wiki-scanner-light`) — an extra slot row
-  holding that position outranks the mode's variant of that slot's agent; **a
-  position with no row is fully governed by the mode.** With `doc-writer` routed
-  and `doc-checker` not, `opus5_only` is entirely live for the checker.
-
-"Not consulted for that slot" and "inert" are different claims, and only the
-first one is true. `orc config list` and `orc config set` NAME the taken bands
-AND the taken positions for exactly that reason: a shadowed setting must never
-be silent, and neither must a partly shadowed one.
-
-Both wiki slots collapse onto `orc-wiki-scanner-opus-5-med` while the mode is on,
-which is why the slot table adds **no agent and no pair**. A slot names the
-POSITION, not the model.
-
-## Tier honesty
-
-A subagent can never outrank the main session. With the mode ON, EVERY dispatch
-needs an Opus 5.5 main session — not just the top executor band. On a lower
-session every role silently falls back to the session model and the tier-honesty
-rule (`return-validation.md`) reports a downgrade on **every** return. That is
-correct behavior, not a bug: hooks cannot block on model, only on effort, so the
-`orc config set opus5_only true` notice is the only up-front warning.
-
-This also means `orc-fast`'s "the orchestrator runs fine at Sonnet medium" holds
-only while the mode is OFF. The effort guard is unchanged — it still matches the
-exact skill name `orc`, and orc-fast is never added to it.
-
-## Trace
-
-Phase 1's `CONFIG` line always records the resolved `opus5_only`, so
-`/orc-retro` can segment per-band outcomes BY dispatch mode.
+# Shared contract — Opus-5-only dispatch mode (hard-gated, forcing)
+
+Canonical rule for routing EVERY dispatched role to Opus 5. Consumed by the full
+orc spine, the ultra lane, and every lane with a fixed-role dispatch; the lint
+pins the `opus5-only.md` pointer into each of them.
+
+## The gate
+
+Read the resolved config at run start (`config.md` defaults ← `orc.config.yaml`).
+The mode is **hard-gated**: absolutely nothing changes unless `opus5_only: true`.
+When it is false (the default), dispatch every role exactly as today — this file
+is inert.
+
+## The mapping (only when `opus5_only: true`)
+
+**Scored executors** — one model, EFFORT as the cost dial, replacing the 5-band
+mixed-model table:
+
+| Score | Executor agent |
+|-----------|---------------------------|
+| `[0,90)`   | `orc-executor-opus-5-low`  |
+| `[90,100]` | `orc-executor-opus-5-med`  |
+
+Two bands, sharing the 90 edge with the default table's top two rows. Since
+v1.0.0 the default table's high end is ALREADY Opus 5.5 with effort as the dial,
+so this mode differs from it only below 41 — a third band here would be a
+distinction the default table stopped making.
+
+**Fixed roles** — dispatch the Opus 5.5 variant **instead of** the default role
+agent: same task slice, same return contract, same phase.
+
+| Role | Default agent | Opus-5-only variant |
+|------|---------------|---------------------|
+| mini executor | `orc-executor-sonnet-5-high` | `orc-executor-opus-5-low` |
+| fast executor | `orc-executor-sonnet-5-med` | `orc-executor-opus-5-low` |
+| mini analyze | `orc-analyze-mini-sonnet-5-high` | `orc-analyze-mini-opus-5-med` |
+| mini plan | `orc-planner-mini-sonnet-5-high` | `orc-planner-mini-opus-5-med` |
+| scout | `orc-scout-sonnet-4-6-high` | `orc-scout-opus-5-low` |
+| pattern codify | `orc-pattern-codifier-sonnet-5-high` | `orc-pattern-codifier-opus-5-med` |
+| wiki scan | `orc-wiki-scanner-sonnet-5-high` (light) | `orc-wiki-scanner-opus-5-low` |
+| retro mine | `orc-retro-sonnet-5-high` | `orc-retro-opus-5-med` |
+
+The eleven roles already pinned to `claude-opus-5-5` — analyst, planner, reviewer,
+verifier, test-author, combiner, learn-writer, advisor, judge, CLAUDE.md writer
+(`orc-claude-writer-opus-5-low`), deep wiki scanner (`orc-wiki-scanner-opus-5-low`)
+— are already
+compliant and dispatch unchanged (their efforts are NOT rewritten by this mode).
+
+**The wiki scan tier ladder (v0.46.0) needs no new pair.**
+Off, the ladder picks deep or light per delta. ON, this mode flips the light tier
+to `orc-wiki-scanner-opus-5-low`. The deep tier is already that agent. So **both
+tiers collapse onto that one shipped agent** and the ladder simply stops applying.
+Do not add a second Opus 5.5 scanner variant. A pair for a tier that cannot occur
+while the flag is on is exactly the phantom this table exists to prevent. Ladder: `../orc-wiki/references/partial-refresh.md`.
+
+## Out of scope — never forced
+
+- **`orc-trace-writer-haiku-4-5` stays Haiku 4.5.** It transcribes a packet the
+  orchestrator hands it — no reasoning, no source reads. It is never in the
+  roster.
+- **`orc-graph-noter-sonnet-4-6-med` is not dispatched at all.** It has no Opus 5.5
+  variant: under the mode a code lane skips graph notes and prints
+  `graph notes: skipped (opus5_only)` (`code-graph.md` §6). The graph's structure
+  layer costs no model, so nothing else about the graph changes.
+- **`orc-diy`.** Its score table is compile-owned (`orc diy compile` →
+  `flow.lock.json`); a DIY flow dispatches whatever its lock says. Re-compile to
+  change it.
+- **`orc-quick` — the ONE exception to the flat precedence below.** Its dispatch
+  gate asks the USER which agent to spawn before every single dispatch, and that
+  hard gate is the lane's entire premise. A forcing mode that collapsed the menu
+  to one option would silently delete it, so the mode is fully INERT there:
+  `opus5_only` is neither read nor honored by `orc-quick`. When the mode is on,
+  the lane SAYS so at the gate (`orc-quick ignores opus5_only — both options are
+  live`), because a shadowed setting must never be silent in either direction.
+- **`/orc-challenge` — UNAFFECTED, not exempt.** All three of its agents are
+  already `claude-opus-5-5` (judge high, advisor medium, reader low), so the mode
+  has nothing to force: zero new pairs, no rename churn, no roster row. The
+  reader's `low` effort is a MEASUREMENT choice, not a cost one — a
+  harder-thinking cold reader reasons around exactly the gaps D4 exists to find
+  — so nothing may ever "upgrade" it.
+  See `orc-quick/references/dispatch-gate.md`.
+
+## It FORCES — precedence is flat while ON
+
+`opus5_only: true` outranks every other dispatch selector:
+
+- each role's default frontmatter pin;
+- **`rubric_bands_override`** — a hand-written executor table is ignored while
+  the mode is on.
+
+Turning the mode off restores both. `rubric_bands` remains granularity
+only, never a preset selector, in both modes.
+
+The forcing is flat across every lane that HONORS the mode — `orc-quick` does
+not, per "Out of scope" above.
+
+### `orc extra` sits ABOVE the mode, and it is not a fourth thing it overrides
+
+Extra decides **whether a Claude agent runs at all**. `opus5_only` and the score
+tables only decide WHICH Claude agent runs where extra did not take it. So the
+mode is neither shadowed nor shadowing — it is simply **NOT CONSULTED** for the
+work extra took:
+
+- **a scored task** — an extra route row covering that score outranks the mode;
+  every score no row covers still resolves on the Opus 5.5 ladder.
+- **a POSITION** (v0.55.0 — `quick-executor`, `fast-executor`, `doc-writer`,
+  `doc-checker`, `wiki-scanner-deep`, `wiki-scanner-light`) — an extra slot row
+  holding that position outranks the mode's variant of that slot's agent; **a
+  position with no row is fully governed by the mode.** With `doc-writer` routed
+  and `doc-checker` not, `opus5_only` is entirely live for the checker.
+
+"Not consulted for that slot" and "inert" are different claims, and only the
+first one is true. `orc config list` and `orc config set` NAME the taken bands
+AND the taken positions for exactly that reason: a shadowed setting must never
+be silent, and neither must a partly shadowed one.
+
+Both wiki slots collapse onto `orc-wiki-scanner-opus-5-low` while the mode is on,
+which is why the slot table adds **no agent and no pair**. A slot names the
+POSITION, not the model.
+
+## Tier honesty
+
+A subagent can never outrank the main session. With the mode ON, EVERY dispatch
+needs an Opus 5.5 main session — not just the top executor band. On a lower
+session every role silently falls back to the session model and the tier-honesty
+rule (`return-validation.md`) reports a downgrade on **every** return. That is
+correct behavior, not a bug: hooks cannot block on model, only on effort, so the
+`orc config set opus5_only true` notice is the only up-front warning.
+
+This also means `orc-fast`'s "the orchestrator runs fine at Sonnet medium" holds
+only while the mode is OFF. The effort guard is unchanged — it still matches the
+exact skill name `orc`, and orc-fast is never added to it.
+
+## Trace
+
+Phase 1's `CONFIG` line always records the resolved `opus5_only`, so
+`/orc-retro` can segment per-band outcomes BY dispatch mode.

package/templates/skills/_shared/phases/review.md

@@ -10,7 +10,7 @@ Emit `PHASE review start`. **Build the slice from `../review-slice.md`** — its
 §1 free check FIRST, then its §2 fields (with the graph on, `graph_changes` keeps
 its rows: no `--brief` here). Superpowers path: its review skill incl. tests
 (Sonnet 4.6 medium), then `../review-slice.md` §3 on its findings — no second
-dispatch. OpenSpec/self path: review worker (Opus 5.5 medium). Pass the resolved
+dispatch. OpenSpec/self path: review worker (Opus 5.5 low). Pass the resolved
 `code_pattern` + its invariants + gate lines for the re-check
 (pattern-gate.md); no resolved pattern → FIRST ask for one (paste/md/none) (H `orc.phase-5.pattern-missing`).
 FE tasks in run → pass `fe_rules[]` from `../../orc-pattern/references/` fe-a11y
@@ -41,7 +41,7 @@ skipped by flow config") — never imply the work was reviewed.
 <!-- diy:when review=on -->
 Dispatch the reviewer exactly as the full lane does — follow the review half
 of `.claude/skills/orc/subskills/orc-review-verify/SKILL.md` (reviewer agent
-`orc-reviewer-opus-5-med`; findings ride the severity ladder from the
+`orc-reviewer-opus-5-low`; findings ride the severity ladder from the
 locked rules, blocking and advisory findings both surfaced). Build its slice
 from `.claude/skills/_shared/review-slice.md` (§1 free check, §3 after-filter).
 <!-- /diy:when -->

package/templates/skills/_shared/phases/trace-verbs.md

@@ -42,7 +42,9 @@ owner is this file has its detail in "Verb detail" below.
 | `JUDGE <analysis\|plan\|implementation> <verdict> round=<n> blocking=<n> advisory=<n> downgraded=<n>` | orc → writer | ultra judgment verdict | `orc/references/ultra-mode.md` |
 | `OUTCOME task=<id> score=<n> band=<range> model=<m> retries=<n> requeues=<n> needs_context=<n> unmet=<n>` | orc → writer | task closed — links the scoring band to what it actually took | `_shared/phases/execution.md` |
 | `FINDING p0=<n> p1=<n> p2=<n> p3=<n>[ pre=<n> suppressed=<n> folded=<n>]` | reviewer→orc → writer | review outcome (P0–P3 severity ladder); the tail counts the after-filter's buckets | `_shared/phases/review.md` |
-| `FINDING-OUTCOME addressed=<n> disputed=<n> wontfix=<n> open=<n> pre=<n> suppressed=<n> :: <cat>:<addressed>/<total>,…` | orc → writer | ONE line at review close — what became of each finding; `orc gotcha quality` reads it | `_shared/phases/trace-verbs.md` |
+| `FINDING-OUTCOME addressed=<n> disputed=<n> wontfix=<n> open=<n> pre=<n> suppressed=<n> :: <cat>:<addressed>/<total>,…` | orc → writer | ONE line at review close (a clean review too) — what became of each finding; `orc gotcha quality` counts it as a review | `_shared/phases/trace-verbs.md` |
+| `REVIEW-WHICH chose=<orc\|project\|skip> name=<the review the rule names> by=<user\|ledger\|learned> :: <file:line of the rule>` | orc → writer | the answer to §0 "which review" when the project names its own review — written ALWAYS, also under habits off; the FINISH nudge reads it | `_shared/phases/trace-verbs.md` |
+| `FIX source=<sonar\|ci\|defect\|pr\|review\|other> introduced_by=<orc\|ai\|human\|unknown> by=<user\|evidence> obs=<id8>[ missed_by=<run>] :: <path:lines> <rule or finding>` | cli | one fix recorded by `orc fix record` (/orc-fix) into the HOST run's trace — the CLI writes it, never a packet | `orc-fix/SKILL.md` |
 | `VERDICT pass\|fail :: <detail>` | verifier→orc → writer | verification outcome | `_shared/phases/verify.md` |
 | `DRIFT loop=<n> :: <user description, compressed>` | orc → writer | mock-example drift-recovery loop opened (hard cap 2 loops) | `_shared/drift-recovery.md` |
 | `TDD-RED task=<id> iter=<n> :: <failing tests>` | executor→orc → writer | TDD repair-loop iteration — the plan's acceptance tests still red | `_shared/phases/trace-verbs.md` |
@@ -170,6 +172,16 @@ reaches the file through a packet — you never append lines by hand.
   the user said "not a problem here" = disputed; declined with no reason = wontfix.
   `pre` and `suppressed` copy the after-filter's buckets; they never gate. The
   `FINDING` tail `pre= suppressed= folded=` is additive (the `sections=` precedent).
+- **`REVIEW-WHICH`** (v2.1.0). ONE line, before ship, in every coding lane that
+  changed code, when the `review-policy` probe says `policy: project`
+  (the review slice, §0). It is written ALWAYS — habits off too — because it is a
+  record of what ran, not a habit. The `ASK any.review.which` line is written in
+  addition only when habits is on. A run that changed code under `policy:
+  project` with no `REVIEW-WHICH` line gets one nudge at FINISH.
+- **`FIX`** (v2.1.0). Written by `orc fix record` ITSELF into the run in flight
+  (`/orc-fix`, a rider that opens no run). A packet that carries it is refused.
+  A lane never narrates it and never repeats it; `orc stats` and `/orc-retro`
+  read it like any other line.
 - **`DRIFT`.** `PHASE mock-example`; canonical `_shared/drift-recovery.md`.
 - **`TDD-RED` / `TDD-GREEN`.** Cap `tdd_loop_max`; a paired TDD task's red proof
   also emits iter=0. `TDD-GREEN` is the non-exempt definition-of-done.
@@ -334,8 +346,10 @@ trace and never emit `SPAWN`/`RETURN`.
 
 ### Files & lifecycle
 
-- Folder: `log_dir` (default `.claude/orc/logs/`). Persistent — **never deleted**
-  (deliberate opposite of the decision log). Top level holds the run `.txt` plus
+- Folder: `log_dir` (default `.claude/orc/logs/`). Persistent (deliberate
+  opposite of the decision log): kept until you run `orc clear logs --apply`, or
+  until the automatic sweep removes it (`log_retention_auto: on`, older than
+  `log_retention_days`). Top level holds the run `.txt` plus
   its sidecars (`.pending.json`, `.jsonl`); generated reports live in
   subfolders (`retro/`).
 - One file per run: **`run-<lane>-<slug>-<DDMMYY>-<HHMMSS>.txt`**, append-only.