@garygentry/feature-forge 0.2.13 → 0.3.0

package/adapters/pi/skills/forge-3-specs/references/stage-exit-protocol.md

@@ -0,0 +1,258 @@
+# Stage Exit Protocol
+
+The single source of truth for how every forge **authoring** stage closes. It
+replaces the old ad-hoc "Next steps:" bullet lists with one fixed, correctly-ordered
+sequence: **verify (if missing or stale) → `/clear` → run the next command.**
+
+Two principles this protocol encodes (do not relitigate — they are locked product
+decisions):
+
+1. **Clearing is recommended on its own merits at every stage boundary** — a clean
+   start for the next stage — *not* as a proxy for a full context window. Window
+   fullness only changes *how emphatically* the clear is recommended, never *whether*
+   it is.
+2. **Verify happens before the clear, never after** — in the authoring session, whether
+   manual **or** auto. Verify's clean-room subagent is dispatched from the *current*
+   session, so the findings digest and any fix decision land where the context to act on
+   them still exists. This holds for auto-verify too: the stage skill dispatches the
+   clean-room verify (and any autoFix) at stage end, in-session, before the exit — it is
+   **not** deferred to the navigator, which runs *after* the `/clear` with none of the
+   authoring context. Clearing first throws that context away.
+
+## How this file is used
+
+The five authoring stages (`forge-0-epic` … `forge-4-backlog`) close with the
+**Scripted Stage Exit**: a short stamped block (below) that runs
+`forge-session.py stage-exit`, obeys the DIRECTIVES it prints per the **directive
+contract** in this file, and prints the script-emitted NEXT-STEPS block verbatim as the
+absolute last output. All the conditional logic the old prose blocks asked the model to
+compute (effective auto-verify, freshness collapse, gate selection, host wording) now
+lives in the script, deterministically; only genuinely interactive work (clean-room
+subagent dispatch, `AskUserQuestion` gates) remains prose — specified once here, not
+per stage.
+
+The loop (`forge-5-loop`) keeps its bespoke exits: it stamps the **standard block**
+(step-6 epic-member handoff) and the **warm variant** (all-done closing) below,
+verbatim. `forge-6-docs` is **terminal** — it stamps no exit block.
+
+A drift-guard test (`tests/test_stage_exit_protocol.py`) asserts each stamp site still
+contains its block, so an edit here must be mirrored into every stamp site (and
+vice-versa).
+
+## Stamp sites
+
+| Stamp site | Block |
+|---|---|
+| `forge-0-epic` … `forge-4-backlog` | scripted-stage-exit stamp |
+| `forge-5-loop` (step-6 epic-member handoff) | standard |
+| `forge-5-loop` (all-done closing → docs) | warm |
+
+The scripted stamp fills one build-time slot, `{stage-exit-args}` — the per-stage
+argument list (e.g. `--feature "{feature}" --stage forge-2-tech`; the epic stage passes
+`--feature "{epic}" --stage forge-0-epic --next-feature "{first-actionable-feature}"`).
+`{feature}` / `{epic}` / `{specsDir}` / `{first-actionable-feature}` remain runtime
+placeholders the skill resolves before running the command, exactly as elsewhere.
+
+<!-- BEGIN: scripted-stage-exit-stamp -->
+**Close this stage with the Scripted Stage Exit** (contract: `references/stage-exit-protocol.md`; do not improvise a "Next steps" list). Run:
+
+```bash
+R="$(bash -c 'for d in "${CLAUDE_PLUGIN_ROOT:-}" "$HOME"/.claude/skills/feature-forge "$HOME"/.claude/plugins/cache/*/feature-forge/* "$HOME"/.claude/plugins/*/feature-forge "$HOME"/.agents/skills/feature-forge ./.agents/skills/feature-forge; do [ -x "$d/scripts/forge-root.sh" ] && exec "$d/scripts/forge-root.sh"; done')"
+[ -n "$R" ] || { echo "feature-forge: cannot locate plugin root" >&2; exit 1; }
+python3 "$R/scripts/forge-session.py" stage-exit {stage-exit-args} --specs-dir "{specsDir}" --host claude
+```
+
+Obey the DIRECTIVES it prints, in order, per the directive contract: `runInStageVerify: true` → dispatch the in-stage clean-room verify now (honoring `autoFixEligible`); `verifyGate: "standard"` → present the Standard Verify Gate; `verifyGate: "manual-print"` → print the `verifyCommand` for the user; non-empty `invalidAutoVerifyKeys` → print a one-line warning. Then **print the NEXT-STEPS block verbatim as your absolute last output — nothing after its sentinel line.**
+<!-- END: scripted-stage-exit-stamp -->
+
+## Directive contract
+
+`stage-exit` emits a DIRECTIVES object and a NEXT-STEPS block. The skill executes the
+directives **in this order**; the script has already computed every conditional, so a
+directive is an instruction, not a question to re-derive.
+
+### `invalidAutoVerifyKeys` (non-empty)
+
+Print a one-line warning first (e.g. "⚠️ forge.config.json `autoVerifyStages` has
+unknown keys: … — they are ignored; fix the typo").
+
+### `runInStageVerify: true` — in-stage auto-verify {stageNoun}
+
+Auto-verify is effective for this stage and verification is outstanding — verify **now,
+in this session** (principle #2 applied to auto-verify: the digest and any fix decision
+land here, where the authoring context still exists — not deferred to a post-`/clear`
+navigator):
+
+1. **Clean-room verify (require-clean).** Dispatch the clean-room `forge-verifier`
+   subagent from this session in require-clean mode — the same path the navigator uses
+   (`skills/forge-verify/SKILL.md`). Dispatch it **synchronously and await its digest
+   inline** — do **not** run it in the background or announce it as "still running";
+   the digest and any fix decision must land in this session. It inherits none of this
+   session's context, so no `/clear` is needed and only a compact digest returns.
+   **Clean-room unavailable** (no `Agent` tool, `forge-verifier` not dispatchable) **or
+   a non-answer returned** (the verifier returned a placeholder / "still running" /
+   delegation message instead of a findings block): do **not** run inline and do **not**
+   silently accept the non-answer as a pass — leave verify **pending** so the navigator
+   catch-up fires on a later Claude-host `/skill:forge`, print the
+   `verifyCommand` for the user to run, and continue to the NEXT-STEPS block.
+2. **Verify passed / no findings** → the fresh verify state is recorded by the
+   clean-room run; continue to the NEXT-STEPS block.
+3. **Verify found findings** →
+   - **`autoFixEligible: true` AND the findings document has zero unresolved decision
+     points** → chain `feature-forge:forge-fix` in-session (it owns its own commit +
+     step tracking), then run a **mandatory re-verify** in require-clean mode. Continue
+     to the NEXT-STEPS block only if the re-verify passes. On any precondition miss, a
+     forge-fix early stop, or a red re-verify, fall through to the digest gate below —
+     never a silent partial mutation. (`autoFixEligible` already folds in the config
+     `autoFix` flag and the clean-tree precondition; a dirty tree or
+     `gitCommitAfterStage: false` arrives here as `false`.)
+   - **`autoFixEligible: false`, or unresolved decision points** → surface a **compact
+     findings digest** as text, then present the gate via `AskUserQuestion`: **Run
+     `forge-fix` now** *(recommended — you are in-context and the digest is right
+     here)* / **Clear + advance anyway** (leave the findings for later) / **Stop
+     here**. Do **not** hard-stop and do **not** silently walk past. Act on the choice,
+     then continue to the NEXT-STEPS block.
+
+### `verifyGate: "standard"` — the Standard Verify Gate
+
+Auto-verify is off for this stage and verification is outstanding (`verifyState` is
+`never`, `stale`, or `failing`). Verify **now, before clearing**, using
+`AskUserQuestion` with exactly these three options — but only when the host has a
+question mechanism **and** the clean-room path is available (the `Agent` tool plus a
+dispatchable `forge-verifier` subagent); otherwise degrade exactly as `manual-print`
+below:
+
+- **Verify {stageNoun} now** *(recommended)* — dispatch the clean-room `forge-verifier`
+  subagent from this session in require-clean mode; the digest returns here so any fix
+  decision keeps its context. One-time — it does **not** change config.
+- **Verify now + enable auto-verify going forward** — verify now **and** patch
+  `"autoVerify": true` into `forge.config.json` in place (preserve formatting and every
+  other key) so future stages verify automatically, no prompt. This complements the
+  `forge-init` opt-in. **Do not auto-commit this config change** — treat it like
+  `notes`: a user-facing edit the user commits on their own cadence, never folded into
+  a stage's artifact commit.
+- **Skip for now** — go straight to the NEXT-STEPS block without verifying. Record this
+  stage's verify status as `"skipped"` in pipeline state (mirroring the existing skip
+  handling) **only** on an explicit skip — a skip does not go stale.
+
+If verify runs and finds findings, handle them exactly as in the in-stage flow above
+(digest + `AskUserQuestion` gate; `autoFixEligible` applies unchanged).
+
+### `verifyGate: "manual-print"`
+
+Verification is outstanding but the host cannot present the gate or dispatch
+clean-room. Do **not** run verify inline — print the `verifyCommand` for the user to
+run (mirroring `autoInvokeNextStage`), offer the auto-verify enable as plain text only
+if a config write is possible, and continue to the NEXT-STEPS block. Verify state stays
+outstanding, so the navigator catch-up can fire later.
+
+### `verifyGate: "none"`
+
+Verification is already resolved (fresh or explicitly skipped) or the in-stage run
+above covers it. Say so in one line and continue to the NEXT-STEPS block.
+
+### `epicReconcile` (epic backflow — present only when there are open requests)
+
+Emitted only when the exiting member carries `open` `epicChangeRequests` (recorded by
+`forge-1-prd`/`forge-2-tech` when the epic *decomposition* itself must change — see
+`references/pipeline-state-schema.json`). Absent on the common path and for standalone
+features. The script has already folded the routing into the NEXT-STEPS block, so this
+directive is informational — you do **not** re-derive the wording:
+
+- `required: true` (at least one `blocksCurrent: true` request) — the NEXT-STEPS block's
+  fenced **primary** command is the epic reconcile command
+  (`/skill:forge-0-epic {epic}`), and the normal next stage is demoted to a
+  follow-up line ("After reconciling, continue with …"). This is *reconcile-before-specs*:
+  proceeding would author artifacts against a decomposition that is about to change. It is
+  strongest when exiting `forge-2-tech` (next is `forge-3-specs`, the point of no cheap
+  return).
+- `reminder: true` (only `blocksCurrent: false` requests) — normal next-stage routing is
+  unchanged; the block appends a non-blocking reminder line ("You also flagged N epic
+  change(s) to reconcile when convenient …"). This is *finish-then-edit*.
+
+Either way the added lines are host-neutral (no literal `/clear`) and sit **above** the
+sentinel; just print the NEXT-STEPS block verbatim as always.
+
+### Deferred decisions — do not solicit next-stage decisions at this exit
+
+Each stage owns its own decisions. At a stage exit, do **not** pull a *later* stage's
+decision forward — do not ask the user (or decide unilaterally) something that properly
+belongs to the next stage's interview (e.g. at `forge-1-prd` exit, don't settle the
+concrete cache backend that `forge-2-tech` will design). Soliciting it here guesses ahead
+of the stage that owns the context, and the answer has nowhere durable to live.
+
+Instead, when you notice a decision that belongs downstream, **record it structurally** as
+a `deferredDecisions[]` entry on this feature's `.pipeline-state.json` (schema in
+`references/pipeline-state-schema.json`; same direct-edit path as `notes` /
+`epicChangeRequests[]`): `question` (phrased for the target stage), optional `rationale`
+and `targetStage`, `raisedBy` (this stage), `raisedAt` (ISO-8601 UTC), `status: "open"`.
+This keeps the exit focused on *this* stage's next-step routing while carrying the open
+question forward for the owning stage to resolve (it flips `status` to `addressed` when it
+does). Prefer a `deferredDecisions[]` entry over stuffing the same thing into the free-text
+`notes` string. This is a recording affordance, not a gate: never block the exit on it.
+
+### The NEXT-STEPS block (always last)
+
+Print the script's NEXT-STEPS block **verbatim as your absolute last output**. Nothing
+follows its final sentinel line (`─ forge: end of stage ─`) — no caveats, no summary,
+no sign-off. The block already carries the `/clear` recommendation (host-aware wording
+via `--host`) and the exact next command, so trailing prose can only push the user's
+next action out of view.
+
+---
+
+## Standard block
+
+Stamped at the loop's step-6 epic-member handoff (finishing feature A → starting
+feature B's PRD). It self-adapts: step 1's verify gate only fires when verification is
+actually outstanding, so at a boundary where verify already ran (or was explicitly
+skipped, or auto-verify is on) it silently collapses to just the `/clear` →
+next-command steps.
+
+Slots: `{stage}` (a lowercase noun phrase), `{verify-command}`, `{next-command}`.
+
+<!-- BEGIN: standard-exit-block -->
+**This stage is done — walk the user through the Stage Exit Protocol** before moving on. The order is fixed, and step 2 is something only the user can do:
+
+1. **Verify {stage} first — if it isn't already verified.** If verify already ran in this session — via the in-stage auto-verify on the authoring stages, or the interactive impl-verify offered above on the loop — or is already fresh on record, or the stage was explicitly skipped, say so and go straight to step 2. Only when `autoVerify` is off for this stage **and** verify is **missing or stale** do you present the **Standard Verify Gate**: verify **now, before clearing**, using `AskUserQuestion` with exactly these three options — but only when the host has a question mechanism **and** the clean-room path is available (the `Agent` tool plus a dispatchable `forge-verifier` subagent):
+   - **Verify {stage} now** *(recommended)* — dispatch the clean-room `forge-verifier` subagent from this session in require-clean mode; the digest returns here so any fix decision keeps its context. One-time — it does **not** change config.
+   - **Verify now + enable auto-verify going forward** — verify now **and** patch `"autoVerify": true` into `forge.config.json` in place (preserve formatting and every other key) so future stages verify automatically, no prompt. This complements the `forge-init` opt-in. **Do not auto-commit this config change** — treat it like `notes`: a user-facing edit the user commits on their own cadence, never folded into a stage's artifact commit.
+   - **Skip for now** — go straight to `/clear` and the next command without verifying. Record this stage's verify status as `"skipped"` in pipeline state (mirroring the existing skip handling) **only** on an explicit skip — a skip does not go stale.
+
+   **Host / clean-room fallback (not a user-selectable option):** if the question mechanism, the `Agent` tool, or the `forge-verifier` subagent is unavailable, do **not** run clean-room — degrade to printing `{verify-command}` for the user to run inline/manually (mirroring `autoInvokeNextStage`), and offer the auto-verify enable as plain text only if a config write is possible.
+2. **Then `/clear`.** Recommended **unconditionally** at this boundary for a clean start — independent of how full the context window is. Every artifact is on disk, so the work survives the clear. **I can't `/clear` for you — you have to run it yourself.**
+3. **Then run the next command** in the fresh session — or re-run `/skill:forge` to let the navigator resume from disk:
+
+   ```
+   {next-command}
+   ```
+<!-- END: standard-exit-block -->
+
+---
+
+## Warm-acceptable variant
+
+Stamp this only at the `forge-5-loop → forge-6-docs` boundary (the all-done result
+report). Here clearing is **optional**: the docs stage benefits from the still-warm
+context of what the loop actually did, and impl-verify is already offered interactively
+by the loop itself, so this block defers rather than re-presenting a gate.
+
+> **Note — no literal `/clear` here.** The warm block lives in `result-reporting.md`, a
+> skill-*own* reference that the adapter build copies **verbatim** (unlike skill bodies,
+> it is not host-term translated), so a literal `/clear` would reach non-Claude adapters
+> undegraded. The warm variant says "clearing is optional" anyway, so it is phrased
+> host-neutrally without the token on purpose — do not reintroduce `/clear` here. (The
+> standard block *does* use `/clear`; that is fine because every standard stamp site is a
+> skill **body**, where `scripts/build-adapters.py` degrades it.)
+
+<!-- BEGIN: warm-exit-block -->
+**The loop is complete — this is the one boundary where clearing before the next stage is optional.**
+
+1. **Verify is already offered above.** Impl-verify is offered interactively right after this report (Step 5b for a standalone feature, Step 6.1 for an epic member) — run it there rather than as a second gate. It runs clean-room, so it needs no fresh session.
+2. **Clearing is optional here — warm is fine.** `forge-6-docs` benefits from the still-warm context of what the loop actually did, so continuing in this same session is the easy default. A cold start also works — every artifact is on disk — but there is no need to force it.
+3. **Then run the next command** — in this warm session, or a fresh one if you prefer:
+
+   ```
+   {next-command}
+   ```
+<!-- END: warm-exit-block -->

package/adapters/pi/skills/forge-4-backlog/SKILL.md

@@ -0,0 +1,175 @@
+---
+# GENERATED — DO NOT EDIT. Source: skills/forge-4-backlog/SKILL.md. Regenerate: python3 scripts/build-adapters.py
+name: forge-4-backlog
+description: Generate a structured backlog.json from forge implementation specs, then validate it via the loop runner. Use when user runs /skill:forge-4-backlog or asks to create a backlog for a forge feature after specs are complete. This is the canonical backlog generator for the forge pipeline. Do NOT trigger for standalone backlog creation outside the forge pipeline context.
+---
+
+# forge-4-backlog — Backlog Generator (pipeline orchestrator)
+
+Generate a complete, validated `backlog.json` from the implementation spec
+suite, ready for the loop runner.
+
+This skill is a **thin orchestrator**: it owns the *pipeline* concerns
+(prerequisite checks, spec loading, plan review, validation, pipeline-state and
+commit). The actual **authoring craft** — granularity, acceptance criteria,
+`agentDelegation`, the schema, examples — lives in the rauf plugin's
+**`author-backlog`** skill, which this skill delegates to. That keeps a single
+home for backlog-authoring knowledge, shared with the repo-wide ad-hoc flow.
+
+## Prerequisites
+
+Read and follow `references/shared-conventions.md` for feature name validation, configuration reading, and force mode handling before proceeding. Resolve the feature directory `{resolvedFeatureDir}` via the **Feature Directory Resolution** block in `references/shared-conventions.md` — do not hardcode `{specsDir}/{feature}/` (see Step 1).
+
+Resolve the **backlog directory** `{backlogDir}`:
+- **`backlogDir` unset (default):** the backlog lives at the resolved feature directory — `{resolvedFeatureDir}/backlog.json` — for both flat and nested features, exactly as today.
+- **`backlogDir` configured:** compose a **per-feature subpath** — `{backlogDir}/{feature}/` — so each epic member's backlog stays independent (the authored file lands at `{backlogDir}/{feature}/backlog.json`). A bare shared `backlogDir` would collide across a multi-feature epic and violate REQ-COMPAT-03; the `{feature}` segment prevents that. Standalone features under a configured `backlogDir` likewise resolve to `{backlogDir}/{feature}/`, which is backward-compatible because each standalone feature name is already unique.
+
+This is the **single** place this rule is implemented. forge-5-loop's backlog-file check must read the same composed `{backlogDir}/{feature}/backlog.json` (that matching forge-5-loop edit lands in item 016), and forge-verify's backlog-mode load uses the same path. rauf itself is unchanged: backlogs remain per-feature and rauf is still launched against a single per-feature backlog path — only the *path composition* changes (REQ-COMPAT-03).
+
+**Let `{resolvedBacklogDir}` denote the composed target of this rule** — i.e. `{backlogDir}/{feature}` when a `backlogDir` is configured, else `{resolvedFeatureDir}`. Every downstream step below (authoring, validation) uses `{resolvedBacklogDir}`, never the bare config value, so the per-feature `{feature}` segment is never dropped.
+
+Resolve the **loop runner** from the `loopRunner` block in `forge.config.json`, filling missing fields from the defaults in `references/forge-config-schema.json` (defaults to rauf). You need its `bin`, `validateCommand`, `versionCommand`, `minRunnerVersion`, and `installHint`.
+
+**Turn structure reminder:** Output analysis/context as text, then route ALL questions through `AskUserQuestion`. Never embed questions in text output — the user will not be prompted and the session will stall.
+
+## Step 1: Validate Prerequisites
+
+**Resolve the feature directory first.** Invoke the **Feature Directory Resolution** block in `references/shared-conventions.md` to turn the bare feature name into `{resolvedFeatureDir}` (exit 0 → stdout is the absolute dir; exit ≥ 1 → STOP and surface the finding verbatim). Read state and specs from `{resolvedFeatureDir}/` everywhere this skill previously wrote `{specsDir}/{feature}/`. Standalone features resolve to their flat path exactly as today.
+
+**Prerequisite check:** Read `{resolvedFeatureDir}/.pipeline-state.json`. If not in force mode, stages `forge-1-prd`, `forge-2-tech`, and `forge-3-specs` must all be `complete`. If not, STOP and tell the user which prerequisites are missing.
+
+After the prerequisite check, invoke the **Stage-Entry Guard** block in `references/shared-conventions.md` with `{stage}` = `forge-4-backlog` — it detects an interrupted or complete `backlog.json`, runs the resume/restart or new-version gate, and stamps entry before Step 2 loads the specs. (The backlog is a single artifact, so "resume" means: reuse the existing `backlog.json` if the previous run wrote it, rather than re-authoring from scratch.)
+
+Then invoke the **Epic-Member Base Guard** block in `references/shared-conventions.md` (this stage does not run Epic Context Injection, so invoke it explicitly here). It self-gates to a no-op for standalone features; for a nested epic member on a branch that lacks the epic manifest it stops with a home-branch pointer (Issue #125).
+
+**Verification check.** Check whether the specs have been verified. If not, use `AskUserQuestion` to warn with the cost of skipping: "Specs haven't been verified yet. Recommended: run `/skill:forge-verify {feature}` first — unverified specs can carry gaps or contradictions that get baked into backlog items and only surface mid-loop, where they're far more expensive to fix. Continue anyway?" Offer **Verify first (recommended)** · **Continue without verifying**.
+
+## Step 2: Load All Specs
+
+Read all spec documents into context:
+- `{resolvedFeatureDir}/PRD.md`
+- `{resolvedFeatureDir}/tech-spec.md`
+- `{resolvedFeatureDir}/##-*.md` (all implementation specs)
+
+If the spec suite is large (8+ documents), focus on loading the architecture layout (01-*), shared types (00-*), and testing strategy documents first. Load individual subsystem specs as needed when writing the corresponding backlog items, rather than loading all specs simultaneously.
+
+## Step 3: Plan the Backlog
+
+Before writing any JSON, walk the specs and create a backlog plan: discrete work items, ordered by dependency (foundation first), with priorities, each scoped for a single loop iteration.
+
+Present the plan as a numbered list:
+```
+Proposed backlog for {feature} ({N} items):
+
+  001 [P1] Scaffold module with project manifest, build config, and entry points
+      Depends on: (none)
+      Specs: 00-core-definitions.md, 01-architecture-layout.md
+
+  002 [P1] Implement shared types and error hierarchy
+      Depends on: 001
+      Specs: 00-core-definitions.md
+  ...
+```
+
+After presenting the plan as text, use `AskUserQuestion` following the **Decision Support** protocol in `references/shared-conventions.md`: recommend this breakdown as the default (it's your evidence-backed read of the specs and dependency order) and name the trade-off that governs item granularity — finer items are each easier to verify in one loop iteration but multiply coordination and dependency edges; coarser items mean fewer handoffs but risk an item too big to complete or verify in a single iteration. Lead with: "I recommend this breakdown. Any items to split, merge, or reorder?" Do NOT include this question in your text output. Wait for the user's response before generating the JSON.
+
+## Step 4: Author backlog.json — delegate to `author-backlog`
+
+**Invoke the rauf plugin's `author-backlog` skill** (via the Skill tool) to write
+`{resolvedBacklogDir}/backlog.json`. Pass it:
+
+- the target backlog directory `{resolvedBacklogDir}`,
+- the approved plan from Step 3,
+- the spec context loaded in Step 2,
+- the project's `typeCheckCommand` / `testCommand` (from `forge.config.json`) so acceptance criteria are concrete and runnable.
+
+`author-backlog` owns all item-quality rules (granularity hard limits, self-contained descriptions, acceptance criteria, `agentDelegation`, the correct `type`/`status` enums, `dependsOn`, `specReferences`, the schema source). Do not re-encode them here — follow whatever it produces.
+
+> **If the rauf plugin / `author-backlog` skill is not available:** fall back to
+> authoring inline using the schema source rule (prefer the project's installed
+> `{stateDir}/backlog.schema.json`, else the published `$id`
+> `https://raw.githubusercontent.com/garygentry/rauf/main/schemas/backlog.schema.json`),
+> and tell the user the rauf plugin provides richer authoring guidance.
+
+**Forge-specific item requirements** layered on top of `author-backlog`'s output:
+- `specReferences` must be paths **relative to the project root** (e.g. `specs/auth/00-core-definitions.md`), NOT relative to the backlog file. The validator resolves them from the project root (not from `--specs-dir`, which only gates the check).
+- **Epic members — cross-member shared-state coupling (#144).** When authoring the backlog for an epic member, check whether this feature writes or migrates a file that a *sibling* member's already-shipped tests pin (its `mutatesShared[]` hint in `epic-manifest.json`, or a shared data corpus / generated fixture its specs say it mutates). If so, schedule a **reconciliation item up front** — before the first mutating item — that regenerates or re-pins the sibling's fixture (or updates the sibling test to the new shape), so the sibling suite stays green instead of red-gating every commit mid-loop. This is the authoring counterpart to forge-verify's CHECK-E10.
+- **Generated-artifact freshness vs. `testCommand` `--check` gates (#145).** If the project's `testCommand` gates on staleness of generated artifacts (`<generator> --check`-style sub-commands that fail when a checked-in generated file is out of date), then any item that regenerates **one** such artifact must regenerate **and commit all** the sibling artifacts those `--check` gates depend on — enumerate the whole gated set, not just the artifact the item is "about", or the item passes locally yet red-gates on the stale-generated check. Pass this to `author-backlog` alongside `testCommand` so the full regeneration sequence lands in each affected item; forge-verify's CHECK-B26 flags a partial set or an ungated regenerator after the fact.
+- **Lifecycle-gated artifacts — no test item forcing a forbidden transition (#150).** A test/e2e item whose acceptance asserts a *published / released / approved / human-reviewed* artifact state must either (a) `dependsOn` an explicit, human-gated publish/review item that legitimately produces that state, or (b) assert the state via a **dev-build / fixture path**. A test item must **never** be the only thing forcing a lifecycle transition another item pins the other way (e.g. one item keeps artifact `X` *draft* while a test demands it *published*, with nothing between to publish it) — the loop, unable to publish or obtain a human sign-off, will **fabricate** the provenance. `author-backlog` owns the general rule; forge-verify's CHECK-B27 flags a contradictory pair with no publisher in the dependency closure after the fact.
+
+> **Backlog schema & rauf contract are unchanged (REQ-COMPAT-03).** Epic membership adds **no** fields to backlog items — dependency edges live in the epic manifest, never in any backlog item. The JSON written here is byte-for-byte the same shape as a pre-epic standalone feature's backlog, and rauf is still launched against a single per-feature backlog path. Only the *path composition* changes (the `{feature}` segment in the backlog-directory rule above), not the schema or rauf's CLI surface.
+
+## Step 5: Validate via the loop runner
+
+Validate the generated backlog by running the runner's **validate command**
+(`loopRunner.validateCommand`), rendered with `{resolvedBacklogDir}` and `{specsDir}`
+substituted — the rauf default:
+
+```bash
+rauf backlog validate . --backlog {resolvedBacklogDir} --specs-dir {specsDir} --json
+```
+
+Interpret the result:
+- **exit 0** → valid (warnings allowed). Proceed.
+- **exit 1** → validation findings. Parse `{ valid, findings[] }`, fix the items, re-run. Do NOT present the backlog to the user until it validates.
+- **exit 2** → usage/IO error (unreadable file, bad JSON). Fix and re-run.
+
+> **forge-4 runs before forge-5's install gate**, so the runner may not be set
+> up yet. Degrade gracefully rather than hard-failing:
+> 1. First run `loopRunner.versionCommand` (`rauf version --json`). If the
+>    binary is **missing**, or its version is **< `minRunnerVersion`**
+>    (semver-compare), do NOT block authoring: keep the authored backlog, emit a
+>    loud warning with `loopRunner.installHint`, mark validation as skipped, and
+>    continue to Step 6. forge-5 will enforce the gate before running.
+> 2. If the binary is present and new enough but the project isn't set up
+>    (`validate` reports the project marker missing), likewise warn and continue
+>    — validation will run cleanly once `rauf install .` has been done.
+
+## Step 6: Review with User
+
+Present a summary: total items N, dependency-chain depth, estimated loop iterations (`ceil(pendingItems * loopIterationMultiplier)`). Note whether validation passed or was skipped (runner not yet available).
+
+State that the backlog is ready and invite adjustments before committing — a statement, not a forced gate: "Backlog is ready. Tell me if you want any items split, merged, or reordered; otherwise I'll record state and commit." Proceed to Step 7 unless the user asks for changes.
+
+## Step 7: Update Pipeline State and Commit
+
+Before writing state or running the stage exit, invoke the **Stage-Completion Re-check** block in `references/shared-conventions.md` with `{stage}` = `forge-4-backlog` — a resumed mid-stage continuation must not overwrite a committed `backlog.json` or re-fire a finished exit.
+
+Write pipeline state conforming to `references/pipeline-state-schema.json`. Follow the Git Commit Protocol in `references/shared-conventions.md`.
+
+1. Update `{resolvedFeatureDir}/.pipeline-state.json`:
+   - Record `artifacts` (path to backlog.json)
+   - Set `stages.forge-4-backlog.basedOnVersions` to `{"forge-1-prd": <current version>, "forge-2-tech": <current version>, "forge-3-specs": <current version>}`
+   - Set `currentStage` to `forge-5-loop`
+   - Check downstream stages (`forge-5-loop`, `forge-6-docs`). If any have `basedOnVersions` referencing an older version of `forge-4-backlog`, set their status to `stale`.
+2. **Offer a note — don't force one.** As a statement (not a blocking question), let the user know they can jot anything worth preserving across sessions and you'll store it in the `notes` field. If they volunteer something, store it; otherwise proceed.
+3. If `gitCommitAfterStage` is true, follow the Git Commit Protocol: stage files, attempt commit (marking `stages.forge-4-backlog.status` `complete` with `commitHash: null` in that commit), then record the artifact-commit hash via the protocol's two-commit follow-up (never `--amend`) only on success. If commit fails, leave status as `in-progress`.
+4. If verification was available but the user chose to skip it, record `stages.forge-verify-backlog.status` as `"skipped"` in pipeline state.
+5. **Close with the Stage Exit Protocol** (single-sourced in `references/stage-exit-protocol.md`; do not improvise a "Next steps" list). Lead with the item count ("Backlog complete with {N} items."), then:
+
+**Close this stage with the Scripted Stage Exit** (contract: `references/stage-exit-protocol.md`; do not improvise a "Next steps" list). Run:
+
+```bash
+R="$(bash -c 'for d in "${FEATURE_FORGE_ROOT:-}" "$HOME"/.claude/skills/feature-forge "$HOME"/.claude/plugins/cache/*/feature-forge/* "$HOME"/.claude/plugins/*/feature-forge "$HOME"/.agents/skills/feature-forge ./.agents/skills/feature-forge; do [ -x "$d/scripts/forge-root.sh" ] && exec "$d/scripts/forge-root.sh"; done')"
+[ -n "$R" ] || { echo "feature-forge: cannot locate plugin root" >&2; exit 1; }
+python3 "$R/scripts/forge-session.py" stage-exit --feature "{feature}" --stage forge-4-backlog --specs-dir "{specsDir}" --host pi
+```
+
+Obey the DIRECTIVES it prints, in order, per the directive contract: `runInStageVerify: true` → dispatch the in-stage clean-room verify now (honoring `autoFixEligible`); `verifyGate: "standard"` → present the Standard Verify Gate; `verifyGate: "manual-print"` → print the `verifyCommand` for the user; non-empty `invalidAutoVerifyKeys` → print a one-line warning. Then **print the NEXT-STEPS block verbatim as your absolute last output — nothing after its sentinel line.**
+
+## Gotchas
+
+- The loop runs each item in a FRESH context. Every item description must be self-contained — `author-backlog` enforces this, but double-check Step-3 plan items aren't "same as above."
+- Spec references must be project-root-relative paths that actually exist — the validate command enforces this when `--specs-dir` is passed (resolving them from the project root).
+- Don't present a backlog to the user before it validates (or before you've explicitly recorded that validation was skipped because the runner isn't installed yet).
+
+---
+
+## Host execution notes (Pi)
+
+This Pi bundle preserves Claude's `AskUserQuestion` references because it ships a Pi compatibility extension registering an `AskUserQuestion` tool. On Pi:
+
+- **User input:** use `AskUserQuestion` for genuine user decisions. It supports multiple questions, option descriptions, recommended ordering, multi-select, previews, and free-form Other/custom answers.
+- **Skill dispatch:** Pi uses `/skill:<name>` commands. If you cannot invoke a skill directly, print the exact `/skill:<name> ...` command for the user to run.
+- **Subagents:** this bundle declares its custom agents (`forge-researcher`, `forge-spec-writer`, `forge-verifier`) as package agents. If a `subagent` tool is registered, dispatch one with `{ agent: "forge-verifier", task: "..." }`, or fan several out concurrently with `{ tasks: [{ agent: "forge-spec-writer", task: "..." }, ...] }`. If no `subagent` tool is available, run that step inline yourself.
+- **Background / monitoring:** run long-lived commands in the foreground and report progress as it arrives.