@ikon85/agent-workflow-kit 0.36.5 → 0.38.0

package/.agents/skills/orchestrate-wave/SKILL.md

@@ -1,6 +1,14 @@
 ---
 name: orchestrate-wave
-"description": "Use when the user hands you a whole WAVE / cluster of pre-planned, file-disjoint slices and wants you to ORCHESTRATE building, verifying AND landing it end-to-end — often AFK / \"ultracode\". Triggers: \"orchestriere Welle #N\" (or any wording that delegates wave-landing responsibility to you), \"ultracode diese Welle / diesen Cluster\", or a wave-anchor issue (a cluster/umbrella issue with file-disjoint sub-issues whose specs are already locked) handed over to land. NOT for a single slice (just `implement` it), NOT for finding/clustering a wave (that's `board-to-waves`), NOT for planning specs (that's `grill-with-docs`/`to-issues`)."
+description: >-
+  Use when the user hands you a whole WAVE / cluster of pre-planned, file-disjoint
+  slices and wants you to ORCHESTRATE building, verifying AND landing it end-to-end —
+  often AFK / "ultracode". Triggers: "orchestriere Welle #N" (or any wording that
+  delegates wave-landing responsibility to you), "ultracode diese Welle / diesen
+  Cluster", or a wave-anchor issue (a cluster/umbrella issue with file-disjoint
+  sub-issues whose specs are already locked) handed over to land. NOT for a single
+  slice (just `implement` it), NOT for finding/clustering a wave (that's `board-to-waves`),
+  NOT for planning specs (that's `grill-with-docs`/`to-issues`).
 ---
 
 # Orchestrate Wave
@@ -25,9 +33,8 @@ node scripts/readiness.mjs check --skill orchestrate-wave --json
 
 - `ready`: continue silently with the required tracker/board context and the
   active `projectRecipe` block.
-- `degraded`: required tracker and managed-board evidence is ready, so keep the
-  complete generic orchestration fallback active, omit only `projectRecipe`,
-  and emit exactly one concise summary: `Readiness degraded — inactive block
+- `degraded`: use generic fallback without `projectRecipe`, and emit exactly one
+  concise summary: `Readiness degraded — inactive block
   projectRecipe (orchestrateWaveRecipe: <state>); using the generic
   orchestration fallback. Run /setup-workflow, configure
   docs/agents/skills/orchestrate-wave.md, then rerun this skill.`
@@ -134,7 +141,7 @@ replace the ~30-second heartbeat; otherwise the standing heartbeat remains requi
    LOCAL annotated tag, so two sessions racing one wave cannot both win. Either
    `acquired: false` or a work signal you did not create means another session
    owns the wave — **STOP**, report the returned `claim.owner` plus the exact
-   branch/worktree, touch nothing. Local coordination state — never push the tag.
+   branch/worktree, touch nothing. After acquisition, begin the claim-bound teardown receipt with `python3 scripts/worktree-lifecycle/session.py begin --anchor <anchor> --owner <claim-owner> --base <wave-HEAD>`; claim and receipt stay local — never push them.
 4. **Wave worktree**: reuse the planning worktree (never re-create; the handoff
    points there). Bring its branch to current `main`:
    `git -C <wave> merge --ff-only origin/main` (if your repo guards destructive
@@ -154,8 +161,7 @@ preflight clean + this run's local claim planted · wave branch ff'd to
 
 Phase 1 uses the selected orchestration mechanics. Resolve every named component,
 produce the **FILE → SLICES** table and overlap graph, then build the conflict hub
-before its dependents. Cut fully file-disjoint waves; shared imports are safe, but
-files edited by multiple slices serialize across waves.
+before its dependents. Cut fully file-disjoint waves; shared imports are safe, but files edited by multiple slices serialize across waves.
 - **Native blocking edges are the frontier authority.** Read the anchor's
   buildable frontier from the tracker's native issue dependencies:
   `python3 scripts/board-sync.py frontier <anchor#>` → `FREI` / `BLOCKED by #…` /
@@ -169,6 +175,7 @@ files edited by multiple slices serialize across waves.
   registries that read targets must serialize helper-owning slices through
   dependency edges, each appending only its own existing artifact after creation.
   Both preserve one owner per shared edit and the no-conflict invariant.
+- **Producer/measurer coupling is a semantic dependency even when files are disjoint.** When one slice mutates a surface and another measures/baselines that surface in the same wave, either order the measurer strictly after the producer through an explicit edge so its baseline reflects post-mutation state, or assign an explicit baseline reconciliation owner at integration to drop producer-resolved entries from the measurer baseline.
 - **Retirement slices require a valid topological deletion order.** Before
   dispatching slices that delete a legacy cluster, map every to-delete module's
   production importers and build the cluster's internal import graph. Order the
@@ -179,7 +186,7 @@ files edited by multiple slices serialize across waves.
 
 **Done when:** FILE→SLICES table exists · each shared file has either one
 declaration-only owner with verbatim consume-only dependents, or an explicit serialized
-owner sequence for eager/validated additions where each owner appends only its own existing artifact · disjoint waves cut in dependency order.
+owner sequence for eager/validated additions where each owner appends only its own existing artifact · every producer/measurer pair has an explicit edge/order or baseline reconciliation owner · disjoint waves cut in dependency order.
 
 ## Phase 2 — Dispatch one wave in parallel (isolated worktree per implementer)
 
@@ -188,8 +195,8 @@ disjoint wave, conflict hub first. Dispatch only `FREI` slices: re-read
 `frontier <anchor#>` before each wave and clear changed edges only via `dep-remove`.
 Before each slice, bind **(a) inline vs delegate** and **(b) tier + effort** under
 Standing rules. Tiny mechanical work may stay inline.
-For Path A/B, create one worktree per agent from wave HEAD, using `§Setup` or
-`git worktree add <path> -b feat/<anchor>-<slug>`.
+For Path A/B, create each new worktree through the active receipt: `python3 scripts/worktree-lifecycle/session.py create --anchor <anchor> --owner <claim-owner> --base <wave-HEAD> --profile <profile> <slice-issue> <slug> <type>`.
+It refuses reused/pre-existing targets; never replace this ownership proof with raw `git worktree add`.
 - **Claim each slice issue at builder launch — the wave claim is not a slice
   claim.** It guards the anchor on this machine only; the slice issue is the
   grabbable unit a second machine or cloud agent sees. Before dispatch, skip and
@@ -236,6 +243,7 @@ A single browser + single dev DB ⇒ verify is serial and yours. **A subagent's
 (real case: "gate PASS" while a size gate was red; files declared missing that
 existed). Never merge on the subagent's word.
 
+- When builders authored or changed browser specs, treat each as an unverified artifact: if `§Verify Recipe` names a local e2e runner, give them ONE central green run with it before the central CI/verify gate. Never invent a command under generic fallback.
 - **Re-run your project's full CI/verify gate CENTRALLY yourself** (`§Verify
   Recipe`). On an integrated verify/coordinator branch (no per-slice issue number)
   a branch-name-derived guard can BLOCK with no matching baseline — a branch-naming
@@ -350,8 +358,8 @@ written · merge order documented.
 - A skill edited during the wave → sync its dual-surface mirror in the SAME PR
   using the tool named by `§Landing` when present; mirror parity remains a pre-PR
   gate.
-- Leave slice worktrees for the user to inspect, or note they're
-  post-merge-removable.
+- After reading every `ANNAHMEN.md`, seal the receipt, inspect it against fetched canonical `main`, then run its `teardown`; only ancestry/one-to-one patch-equivalent exact owned OIDs are compare-deleted, while foreign targets, unique/ambiguous content, changed OIDs, dirty/protected worktrees, and open/unknown PR evidence stop.
+- Keep the completed receipt as recovery-OID evidence; only then release this run's wave claim.
 
 **Done when:** no orphan process · this run's wave claim removed · this run's
 unlanded slice claims released · ANNAHMEN propagated ·

package/.agents/skills/setup-workflow/SKILL.md

@@ -72,6 +72,14 @@ supported as legacy v0 and are never rewritten merely to add the marker.
 Setup and update preserve every extension byte-for-byte unless a future,
 declared schema migration explicitly includes it.
 
+Structured extensions whose Core contract requires every named section declare
+an `all-sections-filled` activation policy and the exact section denominator in
+the Kit readiness catalog. Runtime inspection and readiness both consume that
+same policy, including for a preserved legacy Consumer file. Headings, HTML
+comments, empty fences, placeholders, markers, and explanatory prose outside
+the governed sections do not activate it. Existing v1 extensions without a
+Core activation policy retain their ordinary non-empty-file behavior.
+
 Use an extension for Project language, commands, policy, and capability data
 that the Core skill already knows how to apply. It is additive and cannot
 weaken Core user gates, safety, ownership, or validation. Changes to parsing,
@@ -195,12 +203,16 @@ Worktree Lifecycle for this repository?"* Offer exactly **Yes**, **Later**, and
 
 - **Yes** — create or deepen only the `worktreeLifecycle` section in
   `docs/agents/workflow-capabilities.json`, preserve every other section and
-  unknown key, and reconcile only the exact kit-owned hook commands listed in
-  the seed.
+  unknown key, reconcile an explicit consumer-derived `scratchPatterns` array,
+  explicitly derive and review the consumer's
+  `wrapup.landingGeneratedArtifactPatterns` array from real landing outputs,
+  and reconcile only the exact kit-owned hook commands listed in the seed.
 - **Later / No** — record the choice in the tracked profile without installing
   hook wiring. Ordinary reruns do not ask again.
 - **Existing** — adopt the current section and wiring without normalizing
-  consumer-owned values.
+  consumer-owned values. Preserve an existing landing-artifact policy
+  byte-for-byte; if it is missing, surface the explicit derivation decision and
+  do not write a default.
 - **Disable** — remove only the exact kit-owned hook commands first, then set
   `enabled: false`; retain the profile, setup policy, and unknown keys.
 
@@ -422,6 +434,12 @@ These are **structured-but-empty** crusts that `/retro` grows; do not ask the us
 
 - `docs/agents/skills/spec-self-critique.md` — seed the 12 per-point headings from [spec-self-critique-seed.md](./spec-self-critique-seed.md) so `/retro` has stable append anchors and `spec-self-critique` finds its project layer (suppressing the "layer absent" warning) without inventing project content.
 - `docs/agents/skills/orchestrate-wave.md` — seed the named `§`-section headings from [orchestrate-wave-seed.md](./orchestrate-wave-seed.md) so the `orchestrate-wave` skill's Phase-0 probe finds its project layer. The sections ship **empty** (the exact commands / tunnel / login can't be guessed) → the skill treats an unfilled seed as "layer absent" and runs its generic fallback until the project fills them. A filled section is a manual project-maintenance edit, not something this run invents.
+- The `orchestrateWaveRecipe` readiness capability declares
+  `all-sections-filled` plus the seven named `§` sections. Every one must
+  contain real project instructions before either the runtime inspector or
+  readiness may activate `projectRecipe`. A comment-only or legacy seed remains
+  an intentional generic fallback even when its setup sentinel is
+  `state=filled`.
 - `docs/conventions/spec-completeness.md` — seed **one valid** `## Self-Critique-Check` block (Trigger/Check/Korrektur) from [spec-completeness-seed.md](./spec-completeness-seed.md). A convention file *without* a valid block makes `spec-self-critique` point 8 warn — an empty file is worse than none.
 
 > **Handoff drift-guard (`.claude/hooks/drift-guard.py`).** The repo ships a PreToolUse hook that blocks a `.handoff/*.md` Write when the linked issue's rooted graph is not execute-ready (it delegates all coherence to `scripts/execute-ready-check.py --mode handoff`). It self-filters to `.handoff/*.md` and fires **only once handoff docs exist** — a freshly scaffolded project carries the guard but has nothing to guard yet (silently inoperative until the first `.handoff/` write). This scaffold only **documents** the interplay; it does **not** build new mechanics. Once the project starts emitting handoffs, writes land in `.handoff/` and the guard activates automatically.
@@ -506,6 +524,9 @@ unknown consumer keys share this profile. Apply the transition and hook
 ownership rules from [worktree-lifecycle.md](./worktree-lifecycle.md),
 [workflow-advisories.md](./workflow-advisories.md), and
 [safety-guardrails.md](./safety-guardrails.md) transactionally.
+The landing-artifact policy is deletion authority: it is always an explicit
+consumer-reviewed setup value, never an update-time default or an inference
+from ignored paths.
 
 For Memory Lifecycle, use only the deterministic setup helper from Section A5.
 Do not copy templates with shell commands or edit the capability profile by

package/.agents/skills/setup-workflow/orchestrate-wave-seed.md

@@ -10,8 +10,9 @@ detail. The skill probes it at runtime (Phase 0); with the sections below
 
 > **Section contract.** The skeleton refers to these exact headings by name.
 > Keep them; fill each with your project's real detail. While a section is empty
-> the skill treats the whole layer as absent (Phase-0 sentinel-stub rule) and
-> falls back to generic instructions — so fill them before relying on the recipe.
+> the Kit Core `all-sections-filled` readiness policy treats the whole layer as
+> absent and the skill falls back to generic instructions — so fill them before
+> relying on the recipe.
 
 ## §Setup
 <!-- Project setup steps Phase 0 runs before verify: DB/tunnel/services the

package/.agents/skills/setup-workflow/worktree-lifecycle.md

@@ -29,6 +29,7 @@ advisory, and safe cleanup policy as one unit.
     "operations": [
       "record-choice",
       "reconcile-profile-enabled",
+      "reconcile-landing-artifact-policy",
       "reconcile-hook-wiring"
     ]
   },
@@ -50,7 +51,8 @@ advisory, and safe cleanup policy as one unit.
     "state": "existing",
     "choice": "yes",
     "operations": [
-      "adopt-existing"
+      "adopt-existing",
+      "reconcile-landing-artifact-policy"
     ]
   },
   {
@@ -79,3 +81,29 @@ The exact kit-owned commands are:
 
 Preserve unrelated settings, hook groups, profile sections, and unknown keys.
 Repeated reconciliation with the same choice is byte-identical.
+
+## Scratch classification and sweep
+
+Reconcile an explicit `scratchPatterns` array when enabling a new profile.
+Derive its glob values from the consumer's ignored planning artefacts; an empty
+array is valid. Existing values are consumer-owned and remain byte-identical on
+adoption or rerun. Core never supplies filename defaults.
+
+Also reconcile an explicit
+`wrapup.landingGeneratedArtifactPatterns` array. Derive candidates from the
+consumer's real landing commands and ignored outputs, show the exact list for
+review, and write it only as part of that explicit setup decision. Never copy
+another repository's values or infer deletion authority from `.gitignore`
+alone. An empty array is valid. Existing configured values remain byte-identical.
+If an existing enabled profile lacks the key, report one actionable setup
+decision and leave the project layer unchanged until the consumer confirms the
+derived list.
+
+Patterns use repository-relative POSIX semantics: `*` stays within one segment,
+while `**` crosses `/`; a leading `**/` also matches the repository root.
+Thus `**/__pycache__/**` covers root and nested caches, while `dist-kit/*` does
+not cover `dist-kit/a/b`.
+
+The shipped read-only inventory is
+`python3 scripts/worktree-lifecycle/cleanup.py sweep`. The same profile powers
+its branch issue extraction and scratch-only cleanup verdicts.

package/.agents/skills/wrapup/SKILL.md

@@ -1,7 +1,18 @@
 ---
 name: wrapup
 disable-model-invocation: true
-"description": "Use ONLY when the user types /wrapup. Session-end \"land & clean\" for a finished feature/fix worktree — merges the open PR, kills the worktree dev server, removes the worktree + local branch, and fast-forwards the main checkout so main is current again, then sweeps merged-branch leftovers (local + stale remote whose PR is merged). If the slice isn't landed yet, it first makes it landable (Step 0): commits a dirty tree (after an .env/secret check), pushes, and opens the PR — reusing one if it already exists. User-triggered only (never auto-invoke, never hook). Aborts hard only on: not in a feature worktree, a detected .env/secret, a rejected push, a conflicting PR, or red (FAILURE) checks."
+description: >-
+  Use ONLY when the user directly types $wrapup or /wrapup. Session-end "land & clean" for a
+  finished feature/fix worktree — merges the open PR,
+  kills the worktree dev server, removes the worktree + local branch, and
+  fast-forwards the main checkout so main is current again, then sweeps
+  merged-branch leftovers (local + stale remote whose PR is merged). If the
+  slice isn't landed yet, it first makes it landable (Step 0): commits a dirty
+  tree (after an .env/secret check), pushes, and opens the PR — reusing one if
+  it already exists. User-triggered only (never auto-invoke, never hook). Aborts
+  hard only on: not in a feature worktree, a detected .env/secret, a rejected
+  push, a conflicting PR, terminal red checks, or checks still pending after
+  the bounded wait budget.
 ---
 
 <!-- project-extension:protocol-v1:start -->
@@ -14,15 +25,15 @@ Project extensions may specialize Project details, but cannot weaken Core user g
 
 # wrapup — land PR & tear down worktree
 
-Trigger: user types `/wrapup` (optionally with a PR number). **Manual only** — `disable-model-invocation: true`, no hook, no auto-invoke.
+Trigger: user makes a direct `$wrapup` or `/wrapup` invocation (optionally with a PR number). **Manual only** — `disable-model-invocation: true`, no hook, no auto-invoke.
 
 ## ⚠ Spec context
 
-The user's `/wrapup` input IS the explicit landing authorization for that run.
+The user's direct `$wrapup` or `/wrapup` input IS the explicit landing authorization for that run.
 It authorizes the normal merge flow whether Prod readiness is ready or degraded;
 it never authorizes the agent to invent or configure a deploy target. Never call
-this skill from a hook, another skill, or autonomously. There is no second merge
-confirmation; the pre-flight hard stops are non-negotiable.
+this skill from a hook or another skill. Natural-language requests, indirect skill
+chaining, and autonomous invocation do not authorize it. There is no second merge confirmation; the pre-flight hard stops are non-negotiable.
 
 ## Execution model — script does mechanics, the agent does judgment
 
@@ -52,12 +63,12 @@ conflicting instruction surfaces; never choose a target on the user's behalf.
 
 ### 2 · Retro gate (blocking, optional retro-exit — before anything is committed)
 One reminder, not a merge confirmation:
-> "Already ran a retro? **(a)** yes / continue → landing now. **(b)** you want one first → the retro starts now; afterwards, invoke `/wrapup` again — repo-file patches then travel in this PR."
+> "Already ran a retro? **(a)** yes / continue → landing now. **(b)** you want one first → the retro starts now; afterwards, invoke `$wrapup` or `/wrapup` again — repo-file patches then travel in this PR."
 
 (b) → **invoke the `retro` skill immediately in this run.** Retro is
 model-invocable and non-deploying, and every mutation still has its own approval
 gate. After retro finishes, **land nothing in this run**. Require a **fresh
-explicit `/wrapup` invocation** because retro may have changed the exact diff
+explicit `$wrapup` or `/wrapup` invocation** because retro may have changed the exact diff
 that the next merge authorization covers.
 
 General chaining rule: automatically chain only into a **model-invocable**,
@@ -90,10 +101,32 @@ Run **from the main tree** (the script refuses inside the worktree — an in-wor
 ```bash
 python3 scripts/wrapup-land.py land --branch "<branch>" --title "<title>" --body-file /tmp/wrapup-pr-body.md
 ```
-One call covers: push → PR create/reuse (+ drift markers merged into the body) → `pr-body-check.py` gate (exit 1 = STOP, exit 2 = fail-open warning) → merge gate (red check / `CONFLICTING` = STOP; fresh-PR `UNKNOWN` mergeable passes — the merge itself is the real gate) → **merge** (`--merge` + `--delete-branch`, verified `MERGED`) → dev-server kill (`.dev-ports` ports + cwd-under-worktree walk, own shell ancestry excluded) → worktree remove (no `--force`; refusal = STOP, check surviving processes first) → main `--ff-only` pull + `branch -d` → issue-close verify (auto-close misses are closed manually) → local merged-branch sweep (`-d` only — squash/rebase-merged branches stay a manual call by design) → remote merged-PR sweep (opt-in `wrapup.remoteBranchSweep` in the board profile; PR-status-authoritative via `ls-remote`; deleted remote branches are restorable from the PR page) → anchor-sync (dry-run diff in the report) + anchor completeness check + `execute-ready-check --mode audit` → **upward propagation:** if the anchor's native parent is a Program-PRD, `program-sync` refreshes its Wellenplan (Status + Issue cells) and checks off mechanically completed Phasen-Gates — the slice event is visible at the program level, not only in the wave (`program_sync` block in the report; skipped when the parent isn't a program).
+One call covers: push → PR create/reuse (+ drift markers merged into the body) → `pr-body-check.py` gate (exit 1 = STOP, exit 2 = fail-open warning) → merge gate (pending/null-conclusion checks poll for up to 20 minutes with progress on stderr; terminal red / `CONFLICTING` / timeout = STOP; known zero-step billing or runner failures are named `infrastructure failure`; an already-`MERGED` PR resumes at teardown) → **merge** (`--merge` + `--delete-branch`, verified `MERGED`) → dev-server kill (`.dev-ports` ports + cwd-under-worktree walk, own shell ancestry excluded) → worktree remove (no `--force`; refusal = STOP, check surviving processes first) → main `--ff-only` pull + `branch -d` → issue-close verify (auto-close misses are closed manually) → local merged-branch sweep (`-d` only — squash/rebase-merged branches stay a manual call by design) → remote merged-PR sweep (opt-in `wrapup.remoteBranchSweep` in the board profile; PR-status-authoritative via `ls-remote`; deleted remote branches are restorable from the PR page) → anchor-sync (dry-run diff in the report) + anchor completeness check + `execute-ready-check --mode audit` → **upward propagation:** if the anchor's native parent is a Program-PRD, `program-sync` refreshes its Wellenplan (Status + Issue cells) and checks off mechanically completed Phasen-Gates — the slice event is visible at the program level, not only in the wave (`program_sync` block in the report; skipped when the parent isn't a program).
 
 STOP → diagnose in the main conversation, fix, re-run `land` (an already-merged PR resumes at teardown).
 
+Before merge, the committed worktree policy only nominates exact landing
+evidence. After merge, cleanup reloads `docs/agents/workflow-capabilities.json`
+from canonical `origin/main`; only the identical merged, attempt-bound scratch
+and generator policy authorizes each evidence path and deletion. A mismatch is
+a hard STOP that preserves every file—never bypass it with the unmerged branch
+profile.
+
+Landing provenance has one explicit relinquish route. If the STOP names an
+unfinished or drifted landing attempt and the ambiguous files cannot be restored
+to their frozen identities, run:
+
+```bash
+python3 scripts/wrapup-land.py land --branch "<branch>" --abandon-unfinished-attempt
+```
+
+This archives the started or frozen attempt receipt only. It deletes and claims
+no files, works even when the older creation baseline is missing, and returns
+before push, PR, or merge. The next `land` conservatively protects every current
+matching path; classify or move those blockers first. Do not use relinquish for
+an exact unchanged frozen attempt: a normal retry validates that evidence and
+resumes deterministically.
+
 ### 6 · Post-merge (agent)
 - **Sibling propagation:** for each `drift_markers` entry in the land report, append the note to the target issue's `vor_bau` section + re-stamp its `plan_revision`. Log-based markers → **write directly, then show what was written where** (mandatory report — visibility moved from a pre-gate into the report, decision 2026-07-06); fallback candidates the user hasn't confirmed yet → confirm first. Program context widens the target set to unbuilt wave-stubs/leaves and the Program-PRD itself — same append-only mechanism. **Exception:** appends to the Program-PRD or unbuilt wave-stubs do **not** re-stamp `plan_revision` — that stays reserved for structural wave-plan edits via the `to-waves` escalation path; a mere drift note must not stale-block published stubs.
 - **Anchor close:** report says `anchor_complete: true` → `gh issue close <anchor> -c "Wave complete — all slices merged via PR #<pr>."` and verify board status Done. The guard keeps anchors away from every auto-close — this verified close is the only close path; without it the anchor stays silently open after the last slice. **Then re-run the upward propagation**: the land-time `program-sync` ran BEFORE this close, so on a wave-completing slice the Wellenplan still shows 🔄 and the Phasen-Gate stays unchecked — after the board shows Done, run `python3 scripts/board-sync.py program-sync <program-prd#>` once more (the report's `program_sync.program` names it; skip when the report says the parent is not a program). Board auto-rules can lag the close (Close→Done race) — verify Done first, that's what the token reads.
@@ -108,5 +141,5 @@ STOP → diagnose in the main conversation, fix, re-run `land` (an already-merge
 <!-- readiness:end -->
 
 ## Out of scope
-- Live-verify / DoD: must happen **before** `/wrapup` — this skill lands, it does not verify.
+- Live-verify / DoD: must happen **before** `$wrapup` or `/wrapup` — this skill lands, it does not verify.
 - Other worktrees / their servers stay untouched.

package/.claude/hooks/migration-snapshot-reminder.py

@@ -12,7 +12,7 @@ def main() -> int:
         payload = json.load(sys.stdin)
         core = load_workflow_advisories_core()
         profile = core.load_profile(Path("docs/agents/workflow-capabilities.json"))
-        decision = core.migration_reminder_decision(profile, payload)
+        decision = core.migration_reminder_decision(profile, payload, Path.cwd())
         if decision.context:
             print(json.dumps(hook_event_output(decision.event_name, decision.context)))
     except Exception:

package/.claude/skills/orchestrate-wave/SKILL.md

@@ -33,9 +33,8 @@ node scripts/readiness.mjs check --skill orchestrate-wave --json
 
 - `ready`: continue silently with the required tracker/board context and the
   active `projectRecipe` block.
-- `degraded`: required tracker and managed-board evidence is ready, so keep the
-  complete generic orchestration fallback active, omit only `projectRecipe`,
-  and emit exactly one concise summary: `Readiness degraded — inactive block
+- `degraded`: use generic fallback without `projectRecipe`, and emit exactly one
+  concise summary: `Readiness degraded — inactive block
   projectRecipe (orchestrateWaveRecipe: <state>); using the generic
   orchestration fallback. Run /setup-workflow, configure
   docs/agents/skills/orchestrate-wave.md, then rerun this skill.`
@@ -142,7 +141,7 @@ replace the ~30-second heartbeat; otherwise the standing heartbeat remains requi
    LOCAL annotated tag, so two sessions racing one wave cannot both win. Either
    `acquired: false` or a work signal you did not create means another session
    owns the wave — **STOP**, report the returned `claim.owner` plus the exact
-   branch/worktree, touch nothing. Local coordination state — never push the tag.
+   branch/worktree, touch nothing. After acquisition, begin the claim-bound teardown receipt with `python3 scripts/worktree-lifecycle/session.py begin --anchor <anchor> --owner <claim-owner> --base <wave-HEAD>`; claim and receipt stay local — never push them.
 4. **Wave worktree**: reuse the planning worktree (never re-create; the handoff
    points there). Bring its branch to current `main`:
    `git -C <wave> merge --ff-only origin/main` (if your repo guards destructive
@@ -162,8 +161,7 @@ preflight clean + this run's local claim planted · wave branch ff'd to
 
 Phase 1 uses the selected orchestration mechanics. Resolve every named component,
 produce the **FILE → SLICES** table and overlap graph, then build the conflict hub
-before its dependents. Cut fully file-disjoint waves; shared imports are safe, but
-files edited by multiple slices serialize across waves.
+before its dependents. Cut fully file-disjoint waves; shared imports are safe, but files edited by multiple slices serialize across waves.
 - **Native blocking edges are the frontier authority.** Read the anchor's
   buildable frontier from the tracker's native issue dependencies:
   `python3 scripts/board-sync.py frontier <anchor#>` → `FREI` / `BLOCKED by #…` /
@@ -177,6 +175,7 @@ files edited by multiple slices serialize across waves.
   registries that read targets must serialize helper-owning slices through
   dependency edges, each appending only its own existing artifact after creation.
   Both preserve one owner per shared edit and the no-conflict invariant.
+- **Producer/measurer coupling is a semantic dependency even when files are disjoint.** When one slice mutates a surface and another measures/baselines that surface in the same wave, either order the measurer strictly after the producer through an explicit edge so its baseline reflects post-mutation state, or assign an explicit baseline reconciliation owner at integration to drop producer-resolved entries from the measurer baseline.
 - **Retirement slices require a valid topological deletion order.** Before
   dispatching slices that delete a legacy cluster, map every to-delete module's
   production importers and build the cluster's internal import graph. Order the
@@ -187,7 +186,7 @@ files edited by multiple slices serialize across waves.
 
 **Done when:** FILE→SLICES table exists · each shared file has either one
 declaration-only owner with verbatim consume-only dependents, or an explicit serialized
-owner sequence for eager/validated additions where each owner appends only its own existing artifact · disjoint waves cut in dependency order.
+owner sequence for eager/validated additions where each owner appends only its own existing artifact · every producer/measurer pair has an explicit edge/order or baseline reconciliation owner · disjoint waves cut in dependency order.
 
 ## Phase 2 — Dispatch one wave in parallel (isolated worktree per implementer)
 
@@ -196,8 +195,8 @@ disjoint wave, conflict hub first. Dispatch only `FREI` slices: re-read
 `frontier <anchor#>` before each wave and clear changed edges only via `dep-remove`.
 Before each slice, bind **(a) inline vs delegate** and **(b) tier + effort** under
 Standing rules. Tiny mechanical work may stay inline.
-For Path A/B, create one worktree per agent from wave HEAD, using `§Setup` or
-`git worktree add <path> -b feat/<anchor>-<slug>`.
+For Path A/B, create each new worktree through the active receipt: `python3 scripts/worktree-lifecycle/session.py create --anchor <anchor> --owner <claim-owner> --base <wave-HEAD> --profile <profile> <slice-issue> <slug> <type>`.
+It refuses reused/pre-existing targets; never replace this ownership proof with raw `git worktree add`.
 - **Claim each slice issue at builder launch — the wave claim is not a slice
   claim.** It guards the anchor on this machine only; the slice issue is the
   grabbable unit a second machine or cloud agent sees. Before dispatch, skip and
@@ -244,6 +243,7 @@ A single browser + single dev DB ⇒ verify is serial and yours. **A subagent's
 (real case: "gate PASS" while a size gate was red; files declared missing that
 existed). Never merge on the subagent's word.
 
+- When builders authored or changed browser specs, treat each as an unverified artifact: if `§Verify Recipe` names a local e2e runner, give them ONE central green run with it before the central CI/verify gate. Never invent a command under generic fallback.
 - **Re-run your project's full CI/verify gate CENTRALLY yourself** (`§Verify
   Recipe`). On an integrated verify/coordinator branch (no per-slice issue number)
   a branch-name-derived guard can BLOCK with no matching baseline — a branch-naming
@@ -358,8 +358,8 @@ written · merge order documented.
 - A skill edited during the wave → sync its dual-surface mirror in the SAME PR
   using the tool named by `§Landing` when present; mirror parity remains a pre-PR
   gate.
-- Leave slice worktrees for the user to inspect, or note they're
-  post-merge-removable.
+- After reading every `ANNAHMEN.md`, seal the receipt, inspect it against fetched canonical `main`, then run its `teardown`; only ancestry/one-to-one patch-equivalent exact owned OIDs are compare-deleted, while foreign targets, unique/ambiguous content, changed OIDs, dirty/protected worktrees, and open/unknown PR evidence stop.
+- Keep the completed receipt as recovery-OID evidence; only then release this run's wave claim.
 
 **Done when:** no orphan process · this run's wave claim removed · this run's
 unlanded slice claims released · ANNAHMEN propagated ·

package/.claude/skills/setup-workflow/SKILL.md

@@ -72,6 +72,14 @@ supported as legacy v0 and are never rewritten merely to add the marker.
 Setup and update preserve every extension byte-for-byte unless a future,
 declared schema migration explicitly includes it.
 
+Structured extensions whose Core contract requires every named section declare
+an `all-sections-filled` activation policy and the exact section denominator in
+the Kit readiness catalog. Runtime inspection and readiness both consume that
+same policy, including for a preserved legacy Consumer file. Headings, HTML
+comments, empty fences, placeholders, markers, and explanatory prose outside
+the governed sections do not activate it. Existing v1 extensions without a
+Core activation policy retain their ordinary non-empty-file behavior.
+
 Use an extension for Project language, commands, policy, and capability data
 that the Core skill already knows how to apply. It is additive and cannot
 weaken Core user gates, safety, ownership, or validation. Changes to parsing,
@@ -195,12 +203,16 @@ Worktree Lifecycle for this repository?"* Offer exactly **Yes**, **Later**, and
 
 - **Yes** — create or deepen only the `worktreeLifecycle` section in
   `docs/agents/workflow-capabilities.json`, preserve every other section and
-  unknown key, and reconcile only the exact kit-owned hook commands listed in
-  the seed.
+  unknown key, reconcile an explicit consumer-derived `scratchPatterns` array,
+  explicitly derive and review the consumer's
+  `wrapup.landingGeneratedArtifactPatterns` array from real landing outputs,
+  and reconcile only the exact kit-owned hook commands listed in the seed.
 - **Later / No** — record the choice in the tracked profile without installing
   hook wiring. Ordinary reruns do not ask again.
 - **Existing** — adopt the current section and wiring without normalizing
-  consumer-owned values.
+  consumer-owned values. Preserve an existing landing-artifact policy
+  byte-for-byte; if it is missing, surface the explicit derivation decision and
+  do not write a default.
 - **Disable** — remove only the exact kit-owned hook commands first, then set
   `enabled: false`; retain the profile, setup policy, and unknown keys.
 
@@ -422,6 +434,12 @@ These are **structured-but-empty** crusts that `/retro` grows; do not ask the us
 
 - `docs/agents/skills/spec-self-critique.md` — seed the 12 per-point headings from [spec-self-critique-seed.md](./spec-self-critique-seed.md) so `/retro` has stable append anchors and `spec-self-critique` finds its project layer (suppressing the "layer absent" warning) without inventing project content.
 - `docs/agents/skills/orchestrate-wave.md` — seed the named `§`-section headings from [orchestrate-wave-seed.md](./orchestrate-wave-seed.md) so the `orchestrate-wave` skill's Phase-0 probe finds its project layer. The sections ship **empty** (the exact commands / tunnel / login can't be guessed) → the skill treats an unfilled seed as "layer absent" and runs its generic fallback until the project fills them. A filled section is a manual project-maintenance edit, not something this run invents.
+- The `orchestrateWaveRecipe` readiness capability declares
+  `all-sections-filled` plus the seven named `§` sections. Every one must
+  contain real project instructions before either the runtime inspector or
+  readiness may activate `projectRecipe`. A comment-only or legacy seed remains
+  an intentional generic fallback even when its setup sentinel is
+  `state=filled`.
 - `docs/conventions/spec-completeness.md` — seed **one valid** `## Self-Critique-Check` block (Trigger/Check/Korrektur) from [spec-completeness-seed.md](./spec-completeness-seed.md). A convention file *without* a valid block makes `spec-self-critique` point 8 warn — an empty file is worse than none.
 
 > **Handoff drift-guard (`.claude/hooks/drift-guard.py`).** The repo ships a PreToolUse hook that blocks a `.handoff/*.md` Write when the linked issue's rooted graph is not execute-ready (it delegates all coherence to `scripts/execute-ready-check.py --mode handoff`). It self-filters to `.handoff/*.md` and fires **only once handoff docs exist** — a freshly scaffolded project carries the guard but has nothing to guard yet (silently inoperative until the first `.handoff/` write). This scaffold only **documents** the interplay; it does **not** build new mechanics. Once the project starts emitting handoffs, writes land in `.handoff/` and the guard activates automatically.
@@ -506,6 +524,9 @@ unknown consumer keys share this profile. Apply the transition and hook
 ownership rules from [worktree-lifecycle.md](./worktree-lifecycle.md),
 [workflow-advisories.md](./workflow-advisories.md), and
 [safety-guardrails.md](./safety-guardrails.md) transactionally.
+The landing-artifact policy is deletion authority: it is always an explicit
+consumer-reviewed setup value, never an update-time default or an inference
+from ignored paths.
 
 For Memory Lifecycle, use only the deterministic setup helper from Section A5.
 Do not copy templates with shell commands or edit the capability profile by

package/.claude/skills/setup-workflow/orchestrate-wave-seed.md

@@ -10,8 +10,9 @@ detail. The skill probes it at runtime (Phase 0); with the sections below
 
 > **Section contract.** The skeleton refers to these exact headings by name.
 > Keep them; fill each with your project's real detail. While a section is empty
-> the skill treats the whole layer as absent (Phase-0 sentinel-stub rule) and
-> falls back to generic instructions — so fill them before relying on the recipe.
+> the Kit Core `all-sections-filled` readiness policy treats the whole layer as
+> absent and the skill falls back to generic instructions — so fill them before
+> relying on the recipe.
 
 ## §Setup
 <!-- Project setup steps Phase 0 runs before verify: DB/tunnel/services the

package/.claude/skills/setup-workflow/worktree-lifecycle.md

@@ -29,6 +29,7 @@ advisory, and safe cleanup policy as one unit.
     "operations": [
       "record-choice",
       "reconcile-profile-enabled",
+      "reconcile-landing-artifact-policy",
       "reconcile-hook-wiring"
     ]
   },
@@ -50,7 +51,8 @@ advisory, and safe cleanup policy as one unit.
     "state": "existing",
     "choice": "yes",
     "operations": [
-      "adopt-existing"
+      "adopt-existing",
+      "reconcile-landing-artifact-policy"
     ]
   },
   {
@@ -79,3 +81,29 @@ The exact kit-owned commands are:
 
 Preserve unrelated settings, hook groups, profile sections, and unknown keys.
 Repeated reconciliation with the same choice is byte-identical.
+
+## Scratch classification and sweep
+
+Reconcile an explicit `scratchPatterns` array when enabling a new profile.
+Derive its glob values from the consumer's ignored planning artefacts; an empty
+array is valid. Existing values are consumer-owned and remain byte-identical on
+adoption or rerun. Core never supplies filename defaults.
+
+Also reconcile an explicit
+`wrapup.landingGeneratedArtifactPatterns` array. Derive candidates from the
+consumer's real landing commands and ignored outputs, show the exact list for
+review, and write it only as part of that explicit setup decision. Never copy
+another repository's values or infer deletion authority from `.gitignore`
+alone. An empty array is valid. Existing configured values remain byte-identical.
+If an existing enabled profile lacks the key, report one actionable setup
+decision and leave the project layer unchanged until the consumer confirms the
+derived list.
+
+Patterns use repository-relative POSIX semantics: `*` stays within one segment,
+while `**` crosses `/`; a leading `**/` also matches the repository root.
+Thus `**/__pycache__/**` covers root and nested caches, while `dist-kit/*` does
+not cover `dist-kit/a/b`.
+
+The shipped read-only inventory is
+`python3 scripts/worktree-lifecycle/cleanup.py sweep`. The same profile powers
+its branch issue extraction and scratch-only cleanup verdicts.

package/.claude/skills/skill-manifest.json

@@ -12,7 +12,7 @@
       "projectReleaseProfile": { "evidence": { "type": "json", "paths": ["docs/agents/workflow-capabilities.json"], "validator": "project-release" } },
       "securityAuditRunbook": { "evidence": { "type": "runbook-reference", "paths": ["docs/agents/skills/security-audit.md"], "allowLegacy": true } },
       "prodTarget": { "evidence": { "type": "prod-section", "paths": ["CLAUDE.md", "AGENTS.md"] } },
-      "orchestrateWaveRecipe": { "evidence": { "type": "sentinel", "paths": ["docs/agents/skills/orchestrate-wave.md"], "allowLegacy": true } },
+      "orchestrateWaveRecipe": { "evidence": { "type": "project-extension", "skill": "orchestrate-wave", "paths": ["docs/agents/skills/orchestrate-wave.md"], "activation": { "mode": "all-sections-filled", "sections": ["§Setup", "§Builder Commands", "§Builder Hard Rules", "§Integration Suites", "§Verify Recipe", "§Headless Login", "§Landing"] } } },
       "specCritiqueLayer": { "evidence": { "type": "sentinel", "paths": ["docs/agents/skills/spec-self-critique.md"], "allowLegacy": true } },
       "codeReviewLayer": { "evidence": { "type": "sentinel", "paths": ["docs/agents/code-review.md"], "allowLegacy": true } },
       "verifySpikeLayer": { "evidence": { "type": "sentinel", "paths": ["docs/agents/skills/verify-spike.md"], "allowLegacy": true } },