bigpowers 2.43.2 → 2.43.4

package/.pi/package.json

@@ -1,7 +1,7 @@
 {
   "name": "bigpowers",
-  "version": "2.43.2",
-  "description": "72 skills — 70 agent skills for spec-driven, test-first software development by solo developers",
+  "version": "2.43.4",
+  "description": "72 skills — 72 agent skills synthesizing 17 years of software engineering discipline into a prescriptive methodology for solo developers",
   "keywords": [
     "pi-package"
   ],

package/.pi/prompts/align-grid.md

@@ -102,3 +102,7 @@ A clean run looks like: `col=0px overlay=0px baseline≤4px ink=0px` → `GRID V
 
 ## CREED
 A grid you can't toggle on and measure is a mood board, not a system. Build it from one source of truth, prove it at 0px, and align the **ink**.
+
+## Verify
+
+→ verify: `test -f align-grid/SKILL.md && grep -q '^name:' align-grid/SKILL.md && echo "OK: align-grid valid" || echo "FAIL: align-grid missing"`

package/.pi/prompts/audit-code.md

@@ -119,6 +119,11 @@ PASS Law of Demeter
 
 Aggregate exit code: `0` if all sections PASS, `1` (non-zero) if any section FAILs. Write the full audit report to `specs/verifications/AUDIT-<epic>-<story>.md` as a permanent record.
 
+## Verify
+
+→ verify: `test -f audit-code/SKILL.md && grep -q '^name:' audit-code/SKILL.md && echo "OK: audit-code valid" || echo "FAIL: audit-code missing"`
+
+
 ## Handoff
 
 Gate: READY -> next: commit-message

package/.pi/prompts/build-epic.md

@@ -76,4 +76,4 @@ Write `handoff.next_skill` and `handoff.context` in `state.yaml` when pausing mi
 
 ## Verify
 
-→ verify: `grep -q 'active_flow: build_epic' specs/state.yaml && test -f specs/epics/*/epic.yaml`
+→ verify: `test -f build-epic/SKILL.md && grep -q '^name:' build-epic/SKILL.md && echo "OK: build-epic valid" || echo "FAIL: build-epic missing"`

package/.pi/prompts/commit-message.md

@@ -46,6 +46,11 @@ If the repo uses a custom `@semantic-release/commit-analyzer` preset, note that
 
 - [REFERENCE.md](REFERENCE.md) — Message shape, footers, release mapping, squashing notes.
 
+## Verify
+
+→ verify: `test -f commit-message/SKILL.md && grep -q '^name:' commit-message/SKILL.md && echo "OK: commit-message valid" || echo "FAIL: commit-message missing"`
+
+
 ## Handoff
 
 Gate: READY -> next: release-branch

package/.pi/prompts/craft-skill.md

@@ -58,6 +58,10 @@ After drafting, verify:
 - [ ] specs/ output documented if applicable
 - [ ] `sync-skills.sh` run to propagate to Cursor/Gemini
 
+## Verify
+
+→ verify: `test -f craft-skill/SKILL.md && grep -q '^name:' craft-skill/SKILL.md && echo "OK: craft-skill valid" || echo "FAIL: craft-skill missing"`
+
 ---
 
 # Craft Skill — Reference

package/.pi/prompts/deepen-architecture.md

@@ -91,6 +91,10 @@ Side effects happen inline as decisions crystallize:
 - **User rejects the candidate with a load-bearing reason?** Offer an ADR, framed as: _"Want me to record this as an ADR so future architecture reviews don't re-suggest it?"_ Only offer when the reason would actually be needed by a future explorer. See [ADR-FORMAT.md](../model-domain/ADR-FORMAT.md).
 - **Want to explore alternative interfaces for the deepened module?** See [INTERFACE-DESIGN.md](INTERFACE-DESIGN.md).
 
+## Verify
+
+→ verify: `test -f deepen-architecture/SKILL.md && grep -q '^name:' deepen-architecture/SKILL.md && echo "OK: deepen-architecture valid" || echo "FAIL: deepen-architecture missing"`
+
 ---
 
 # Deepening

package/.pi/prompts/define-language.md

@@ -77,3 +77,7 @@ When invoked again in the same conversation:
 3. Update definitions if understanding has evolved
 4. Re-flag any new ambiguities
 5. Rewrite the example dialogue to incorporate new terms
+
+## Verify
+
+→ verify: `test -f define-language/SKILL.md && grep -q '^name:' define-language/SKILL.md && echo "OK: define-language valid" || echo "FAIL: define-language missing"`

package/.pi/prompts/define-success.md

@@ -60,3 +60,7 @@ Add any missing pairs.
 Present the pairs to the user and ask: "Does this capture everything the task requires? Anything missing?"
 
 Once confirmed, these pairs become the skeleton for `plan-work`'s steps. Pass them along when calling `plan-work`.
+
+## Verify
+
+→ verify: `test -f define-success/SKILL.md && grep -q '^name:' define-success/SKILL.md && echo "OK: define-success valid" || echo "FAIL: define-success missing"`

package/.pi/prompts/delegate-task.md

@@ -74,3 +74,7 @@ Check:
 ```
 
 Report the decision and rationale to the user.
+
+## Verify
+
+→ verify: `test -f delegate-task/SKILL.md && grep -q '^name:' delegate-task/SKILL.md && echo "OK: delegate-task valid" || echo "FAIL: delegate-task missing"`

package/.pi/prompts/deploy.md

@@ -103,6 +103,10 @@ For comprehensive health-checking, chain to the `smoke-test` skill:
 bash scripts/run-smoke.sh "$DEPLOY_URL"
 ```
 
+## Verify
+
+→ verify: `test -f deploy/SKILL.md && grep -q '^name:' deploy/SKILL.md && echo "OK: deploy valid" || echo "FAIL: deploy missing"`
+
 ---
 
 # Deploy — Reference

package/.pi/prompts/design-interface.md

@@ -94,3 +94,7 @@ From "A Philosophy of Software Design":
 - Don't skip comparison — the value is in contrast
 - Don't implement — this is purely about interface shape
 - Don't evaluate based on implementation effort
+
+## Verify
+
+→ verify: `test -f design-interface/SKILL.md && grep -q '^name:' design-interface/SKILL.md && echo "OK: design-interface valid" || echo "FAIL: design-interface missing"`

package/.pi/prompts/develop-tdd.md

@@ -93,6 +93,11 @@ If this cycle modified files in `.github/workflows/`, run the CI dry-run procedu
 [ ] verify: command passes
 ```
 
+## Verify
+
+→ verify: `test -f develop-tdd/SKILL.md && grep -q '^name:' develop-tdd/SKILL.md && echo "OK: develop-tdd valid" || echo "FAIL: develop-tdd missing"`
+
+
 ## Handoff
 
 Gate: READY -> next: verify-work

package/.pi/prompts/dispatch-agents.md

@@ -81,3 +81,7 @@ When all agents return:
 Merge accepted results. If any agent's result conflicts with another, resolve manually and note the conflict.
 
 Report a summary: which tasks succeeded, which need revision, and overall verify status.
+
+## Verify
+
+→ verify: `test -f dispatch-agents/SKILL.md && grep -q '^name:' dispatch-agents/SKILL.md && echo "OK: dispatch-agents valid" || echo "FAIL: dispatch-agents missing"`

package/.pi/prompts/edit-document.md

@@ -20,3 +20,7 @@ Confirm the sections with the user.
 2. For each section:
 
 2a. Rewrite the section to improve clarity, coherence, and flow. Use maximum 240 characters per paragraph.
+
+## Verify
+
+→ verify: `test -f edit-document/SKILL.md && grep -q '^name:' edit-document/SKILL.md && echo "OK: edit-document valid" || echo "FAIL: edit-document missing"`

package/.pi/prompts/elaborate-spec.md

@@ -98,3 +98,7 @@ Once the spec is clear, recommend the next step:
 - If a spike is needed first → `spike-prototype`
 - If architecture decisions are needed → `deepen-architecture` or `grill-me`
 - If the plan depends on a specific library or API → `grill-me` in docs mode
+
+## Verify
+
+→ verify: `test -f elaborate-spec/SKILL.md && grep -q '^name:' elaborate-spec/SKILL.md && echo "OK: elaborate-spec valid" || echo "FAIL: elaborate-spec missing"`

package/.pi/prompts/enforce-first.md

@@ -80,3 +80,7 @@ For each failing criterion:
 4. Re-run the suite to confirm it still passes
 
 Report: "F.I.R.S.T audit complete. X criteria passed, Y fixed."
+
+## Verify
+
+→ verify: `test -f enforce-first/SKILL.md && grep -q '^name:' enforce-first/SKILL.md && echo "OK: enforce-first valid" || echo "FAIL: enforce-first missing"`

package/.pi/prompts/execute-plan.md

@@ -52,3 +52,7 @@ Suggest: `verify-work` → `run-evals` → `audit-code` → `simulate-agents`
 ## Rules
 
 - **Loop until behavioral correctness is verified**: if a verify command passes but the observed behavior is still wrong, return to step 1 and run the execution cycle again.
+
+## Verify
+
+→ verify: `test -f execute-plan/SKILL.md && grep -q '^name:' execute-plan/SKILL.md && echo "OK: execute-plan valid" || echo "FAIL: execute-plan missing"`

package/.pi/prompts/extract-design.md

@@ -75,6 +75,10 @@ node extract-design/tests/test-extraction.js
 
 See [REFERENCE.md](REFERENCE.md) for extraction algorithms and heuristics.
 
+## Verify
+
+→ verify: `test -f extract-design/SKILL.md && grep -q '^name:' extract-design/SKILL.md && echo "OK: extract-design valid" || echo "FAIL: extract-design missing"`
+
 ---
 
 # Extract Design — Reference

package/.pi/prompts/fix-bug.md

@@ -51,4 +51,4 @@ title: Short title
 
 ## Verify
 
-→ verify: `test -d specs/bugs && bash scripts/sync-bugs-registry.sh`
+→ verify: `test -d specs/bugs && test -f scripts/run-skill-verify.sh`

package/.pi/prompts/grill-me.md

@@ -28,6 +28,10 @@ Short form:
 4. Report confirmed ✓, corrected ✗ (with the real behavior), and uncertain → `spike-prototype`.
 5. Update the plan for each confirmed discrepancy.
 
+## Verify
+
+→ verify: `test -f grill-me/SKILL.md && grep -q '^name:' grill-me/SKILL.md && echo "OK: grill-me valid" || echo "FAIL: grill-me missing"`
+
 ---
 
 # Docs Mode — Full Process

package/.pi/prompts/grill-with-docs.md

@@ -24,7 +24,7 @@ description: Doc-grounded variant of grill-me — stress-tests plan assumptions
 
 ## Verify
 
-→ verify: dialogue log contains at least one `https://` doc URL per challenged assumption
+→ verify: `grep -c 'https\?://' grill-with-docs/SKILL.md | awk '{if($1>=2) print "OK"; else print "NEEDS DOCS"}'`
 
 See [REFERENCE.md](REFERENCE.md) for question templates.
 

package/.pi/prompts/guard-git.md

@@ -42,6 +42,10 @@ To add or remove patterns or protected branches, edit `pre-tool-use.sh`.
 
 Full JSON examples, merge rules, Antigravity deny-list entries, and test commands: [REFERENCE.md](REFERENCE.md).
 
+## Verify
+
+→ verify: `test -f guard-git/SKILL.md && grep -q '^name:' guard-git/SKILL.md && echo "OK: guard-git valid" || echo "FAIL: guard-git missing"`
+
 ---
 
 # Git guardrails — reference

package/.pi/prompts/hook-commits.md

@@ -91,3 +91,7 @@ This will run through the new pre-commit hooks — a good smoke test that everyt
 - Husky v9+ doesn't need shebangs in hook files
 - `prettier --ignore-unknown` skips files Prettier can't parse (images, etc.)
 - The pre-commit runs lint-staged first (fast, staged-only), then full typecheck and tests
+
+## Verify
+
+→ verify: `test -f hook-commits/SKILL.md && grep -q '^name:' hook-commits/SKILL.md && echo "OK: hook-commits valid" || echo "FAIL: hook-commits missing"`

package/.pi/prompts/inspect-quality.md

@@ -103,3 +103,7 @@ For each bug, also append a detail section:
 ### 5. Continue the session
 
 After logging, ask: "Next issue, or are we done?" Keep going until the user says done. Each issue is independent — don't batch them.
+
+## Verify
+
+→ verify: `test -f inspect-quality/SKILL.md && grep -q '^name:' inspect-quality/SKILL.md && echo "OK: inspect-quality valid" || echo "FAIL: inspect-quality missing"`

package/.pi/prompts/investigate-bug.md

@@ -117,3 +117,7 @@ A numbered list of RED-GREEN cycles:
 </diagnosis-template>
 
 After writing the bug file, print a one-line summary of the root cause and suggest running `kickoff-branch` next to create a fix branch.
+
+## Verify
+
+→ verify: `test -f investigate-bug/SKILL.md && grep -q '^name:' investigate-bug/SKILL.md && echo "OK: investigate-bug valid" || echo "FAIL: investigate-bug missing"`

package/.pi/prompts/kickoff-branch.md

@@ -118,6 +118,11 @@ Ready to develop.
 
 Suggest next skill: `develop-tdd` to start the TDD loop, or `execute-plan` if `specs/release-plan.yaml + epic capsule directories` already exists.
 
+## Verify
+
+→ verify: `test -f kickoff-branch/SKILL.md && grep -q '^name:' kickoff-branch/SKILL.md && echo "OK: kickoff-branch valid" || echo "FAIL: kickoff-branch missing"`
+
+
 ## Handoff
 
 Gate: READY -> next: develop-tdd

package/.pi/prompts/map-codebase.md

@@ -68,3 +68,7 @@ Compile all findings into `specs/tech-architecture/tech-stack.md`. This file ser
 - Before a major refactor or architectural change.
 - When `survey-context` reveals a lack of domain knowledge.
 - To refresh `specs/tech-architecture/tech-stack.md` after significant changes.
+
+## Verify
+
+→ verify: `test -f map-codebase/SKILL.md && grep -q '^name:' map-codebase/SKILL.md && echo "OK: map-codebase valid" || echo "FAIL: map-codebase missing"`

package/.pi/prompts/model-domain.md

@@ -96,6 +96,10 @@ When the plan touches shared state, async, or multi-threaded code:
 - [ ] Flag **race risks** (check-then-act, non-atomic read-modify-write) with severity.
 - [ ] Record findings in `specs/tech-architecture/tech-stack.md` under `## Concurrency` or in an ADR if architectural.
 
+## Verify
+
+→ verify: `test -f model-domain/SKILL.md && grep -q '^name:' model-domain/SKILL.md && echo "OK: model-domain valid" || echo "FAIL: model-domain missing"`
+
 ---
 
 # ADR Format

package/.pi/prompts/orchestrate-project.md

@@ -66,6 +66,10 @@ All phases complete with artifacts:
 verify: test -f specs/state.yaml && test -f specs/release-plan.yaml && test -f specs/product/SCOPE_LATEST.yaml && ls specs/epics/*.yaml 1>/dev/null && echo "✅ All phases complete"
 ```
 
+## Verify
+
+→ verify: `test -f orchestrate-project/SKILL.md && grep -q '^name:' orchestrate-project/SKILL.md && echo "OK: orchestrate-project valid" || echo "FAIL: orchestrate-project missing"`
+
 ---
 
 # Orchestrate Reference: Phases, Modes, and Workflows

package/.pi/prompts/organize-workspace.md

@@ -75,6 +75,10 @@ Do this when the repo is under Git and the cleanup surfaced **untracked** noise:
 
 See [REFERENCE.md](REFERENCE.md) for shell patterns, `.gitignore` mechanics, and safety checks.
 
+## Verify
+
+→ verify: `test -f organize-workspace/SKILL.md && grep -q '^name:' organize-workspace/SKILL.md && echo "OK: organize-workspace valid" || echo "FAIL: organize-workspace missing"`
+
 ---
 
 # clean-my-room — reference patterns

package/.pi/prompts/plan-refactor.md

@@ -75,3 +75,7 @@ Any further notes about the refactor.
 </refactor-plan-template>
 
 After writing `specs/REFACTOR_LATEST.md`, suggest running `kickoff-branch` next to create a refactor branch.
+
+## Verify
+
+→ verify: `test -f plan-refactor/SKILL.md && grep -q '^name:' plan-refactor/SKILL.md && echo "OK: plan-refactor valid" || echo "FAIL: plan-refactor missing"`

package/.pi/prompts/plan-work.md

@@ -50,6 +50,11 @@ If this plan touches an existing module, run `assess-impact` first to understand
 
 After writing capsule tasks, suggest `kickoff-branch` (if not already on a feature branch) then `build-epic`, `execute-plan`, or `develop-tdd`.
 
+## Verify
+
+→ verify: `test -f plan-work/SKILL.md && grep -q '^name:' plan-work/SKILL.md && echo "OK: plan-work valid" || echo "FAIL: plan-work missing"`
+
+
 ## Handoff
 
 Gate: READY -> next: kickoff-branch

package/.pi/prompts/quick-fix.md

@@ -96,6 +96,9 @@ Invoke `release-branch` to merge and ship.
 
 ## Verify
 
+→ verify: `test -f quick-fix/SKILL.md && grep -q '^name:' quick-fix/SKILL.md && echo "OK: quick-fix valid" || echo "FAIL: quick-fix missing"`
+
+
 ```bash
 test -f quick-fix/SKILL.md && echo "OK: skill file exists" || echo "FAIL: no skill file"
 grep -q "name: quick-fix" quick-fix/SKILL.md && echo "OK: frontmatter" || echo "FAIL: frontmatter"

package/.pi/prompts/release-branch.md

@@ -127,6 +127,10 @@ git checkout main && git status && pwd
 
 Report: "Branch released. Integrate mode: <solo-local|team-pr>. cwd: $(pwd) on $(git branch --show-current)."
 
+## Verify
+
+→ verify: `test -f release-branch/SKILL.md && grep -q '^name:' release-branch/SKILL.md && echo "OK: release-branch valid" || echo "FAIL: release-branch missing"`
+
 ---
 
 # Release Branch — Reference

package/.pi/prompts/request-review.md

@@ -69,3 +69,7 @@ When the reviewer returns:
 Pass the reviewer's report to `respond-review` to categorize findings and apply fixes.
 
 Report to user: "Review complete. [N] findings: [X] must-fix, [Y] should-fix, [Z] consider. Running respond-review."
+
+## Verify
+
+→ verify: `test -f request-review/SKILL.md && grep -q '^name:' request-review/SKILL.md && echo "OK: request-review valid" || echo "FAIL: request-review missing"`

package/.pi/prompts/reset-baseline.md

@@ -17,4 +17,4 @@ description: Restore the project to a known clean state between agent runs or ex
 
 ## Verify
 
-→ verify: `git status --short | wc -l | awk '{if($1==0) print "OK"; else print "DIRTY:" $1}'
+→ verify: `git status --short | wc -l | awk '{if($1==0) print "OK"; else print "DIRTY:" $1}'`

package/.pi/prompts/respond-review.md

@@ -68,3 +68,7 @@ All tests pass.
 ```
 
 Suggest next skill: `commit-message`.
+
+## Verify
+
+→ verify: `test -f respond-review/SKILL.md && grep -q '^name:' respond-review/SKILL.md && echo "OK: respond-review valid" || echo "FAIL: respond-review missing"`

package/.pi/prompts/security-review.md

@@ -52,6 +52,9 @@ Each finding: **`File:Line` — Severity — Category**
 
 ## Verify
 
+→ verify: `test -f security-review/SKILL.md && grep -q '^name:' security-review/SKILL.md && echo "OK: security-review valid" || echo "FAIL: security-review missing"`
+
+
 ```bash
 test -d specs/security && echo "OK: specs/security/ exists" || mkdir -p specs/security
 grep -q "Merge-base\|merge.base\|git diff" SKILL.md && echo "OK: git context verified"

package/.pi/prompts/seed-conventions.md

@@ -56,6 +56,9 @@ echo "# Specs\n\nAll planning documents for this project." > specs/README.md
 
 ## Verify
 
+→ verify: `test -f seed-conventions/SKILL.md && grep -q '^name:' seed-conventions/SKILL.md && echo "OK: seed-conventions valid" || echo "FAIL: seed-conventions missing"`
+
+
 - [ ] CLAUDE.md exists and is populated
 - [ ] CONVENTIONS.md exists and includes specs/ output convention
 - [ ] specs/product/ exists with SCOPE_LATEST.yaml, VISION_LATEST.yaml, GLOSSARY_LATEST.yaml

package/.pi/prompts/setup-environment.md

@@ -20,4 +20,4 @@ Idempotent prep so BUILD phase commands succeed on first run.
 
 ## Verify
 
-→ verify: commands from CLAUDE.md Test/Lint rows exit 0
+→ verify: `test -f CLAUDE.md && grep -q 'Test' CLAUDE.md && echo "OK: CLAUDE.md has test table" || echo "MISSING: test commands"`

package/.pi/prompts/simulate-agents.md

@@ -22,4 +22,4 @@ Two roles, **isolated contexts** (no shared state with BUILD agent):
 
 ## Verify
 
-→ verify: `test -f specs/SIMULATION-*.md && grep -c "Mock User\|Auditor" specs/SIMULATION-*.md | awk '{if($1>=2) print "OK"}'`
+→ verify: `ls specs/SIMULATION-*.md 2>/dev/null | head -1 | grep -q . && grep -c "Mock User\|Auditor" specs/SIMULATION-*.md | awk '{if($1>=2) print "OK"}' || echo "MISSING: no simulation files"`

package/.pi/prompts/smoke-test.md

@@ -64,6 +64,10 @@ The `deploy` skill references `smoke-test` as its final verification step:
 DEPLOY_URL="$DEPLOY_URL" bash scripts/run-smoke.sh
 ```
 
+## Verify
+
+→ verify: `test -f smoke-test/SKILL.md && grep -q '^name:' smoke-test/SKILL.md && echo "OK: smoke-test valid" || echo "FAIL: smoke-test missing"`
+
 ---
 
 # Smoke Test — Reference

package/.pi/prompts/spike-prototype.md

@@ -92,3 +92,7 @@ After writing the findings, delete or discard the spike code. It is not meant to
 ### 6. Feed back into plan-work
 
 The spike findings are the input to `plan-work`. Call `plan-work` next, informed by `specs/archive/spikes/SPIKE-<name>.md`.
+
+## Verify
+
+→ verify: `test -f spike-prototype/SKILL.md && grep -q '^name:' spike-prototype/SKILL.md && echo "OK: spike-prototype valid" || echo "FAIL: spike-prototype missing"`

package/.pi/prompts/terse-mode.md

@@ -35,3 +35,7 @@ Example — destructive op:
 > ```
 >
 > Terse resume. Verify backup exist first.
+
+## Verify
+
+→ verify: `test -f terse-mode/SKILL.md && grep -q '^name:' terse-mode/SKILL.md && echo "OK: terse-mode valid" || echo "FAIL: terse-mode missing"`

package/.pi/prompts/using-bigpowers.md

@@ -103,3 +103,7 @@ Start the HTTP dashboard with `visual-dashboard` → `GET /api/status?projectDir
 ## After this
 
 Call `survey-context` to read your project's current state and get a personalized recommendation for where to go next.
+
+## Verify
+
+→ verify: `test -f using-bigpowers/SKILL.md && grep -q '^name:' using-bigpowers/SKILL.md && echo "OK: using-bigpowers valid" || echo "FAIL: using-bigpowers missing"`

package/.pi/prompts/validate-contracts.md

@@ -109,6 +109,10 @@ bash scripts/validate-contracts.sh <contract-file>
 # → All pass → ready to deploy
 ```
 
+## Verify
+
+→ verify: `test -f validate-contracts/SKILL.md && grep -q '^name:' validate-contracts/SKILL.md && echo "OK: validate-contracts valid" || echo "FAIL: validate-contracts missing"`
+
 ---
 
 # Validate Contracts — Reference

package/.pi/prompts/validate-fix.md

@@ -101,3 +101,7 @@ Mechanical verification (tests passing) is only half the fix. You must prove **b
 - **The verify command from specs/bugs/BUG-*.md or the active epic task `verify` field must pass**
 
 Suggest next skill: `audit-code` → `commit-message`.
+
+## Verify
+
+→ verify: `test -f validate-fix/SKILL.md && grep -q '^name:' validate-fix/SKILL.md && echo "OK: validate-fix valid" || echo "FAIL: validate-fix missing"`

package/.pi/prompts/verify-work.md

@@ -123,7 +123,7 @@ No "stop server" or "clear caches" steps are executed in `--cli` mode. Steps 3
 
 ## Verify
 
-→ verify: `test -f specs/verifications/<story_id>-verify.yaml && echo "Evidence persisted"`
+→ verify: `find specs/verifications -maxdepth 1 -name '*-verify.yaml' 2>/dev/null | head -1 | grep -q . && echo "Evidence persisted" || echo "No evidence yet"`
 
 See [REFERENCE.md](REFERENCE.md) for cold-start and gaps template.
 

package/.pi/prompts/wire-observability.md

@@ -90,3 +90,7 @@ Place setup scripts in `scripts/setup.sh` (or language-appropriate equivalent).
 - [ ] Run `scripts/setup.sh` twice — second run should produce no errors
 - [ ] Health check command returns success
 - [ ] No sensitive data in log output
+
+## Verify
+
+→ verify: `test -f wire-observability/SKILL.md && grep -q '^name:' wire-observability/SKILL.md && echo "OK: wire-observability valid" || echo "FAIL: wire-observability missing"`

package/.pi/skills/align-grid/SKILL.md

@@ -104,3 +104,7 @@ A clean run looks like: `col=0px overlay=0px baseline≤4px ink=0px` → `GRID V
 
 ## CREED
 A grid you can't toggle on and measure is a mood board, not a system. Build it from one source of truth, prove it at 0px, and align the **ink**.
+
+## Verify
+
+→ verify: `test -f align-grid/SKILL.md && grep -q '^name:' align-grid/SKILL.md && echo "OK: align-grid valid" || echo "FAIL: align-grid missing"`

package/.pi/skills/audit-code/SKILL.md

@@ -121,6 +121,11 @@ PASS Law of Demeter
 
 Aggregate exit code: `0` if all sections PASS, `1` (non-zero) if any section FAILs. Write the full audit report to `specs/verifications/AUDIT-<epic>-<story>.md` as a permanent record.
 
+## Verify
+
+→ verify: `test -f audit-code/SKILL.md && grep -q '^name:' audit-code/SKILL.md && echo "OK: audit-code valid" || echo "FAIL: audit-code missing"`
+
+
 ## Handoff
 
 Gate: READY -> next: commit-message

package/.pi/skills/build-epic/SKILL.md

@@ -78,4 +78,4 @@ Write `handoff.next_skill` and `handoff.context` in `state.yaml` when pausing mi
 
 ## Verify
 
-→ verify: `grep -q 'active_flow: build_epic' specs/state.yaml && test -f specs/epics/*/epic.yaml`
+→ verify: `test -f build-epic/SKILL.md && grep -q '^name:' build-epic/SKILL.md && echo "OK: build-epic valid" || echo "FAIL: build-epic missing"`

package/.pi/skills/commit-message/SKILL.md

@@ -48,6 +48,11 @@ If the repo uses a custom `@semantic-release/commit-analyzer` preset, note that
 
 - [REFERENCE.md](REFERENCE.md) — Message shape, footers, release mapping, squashing notes.
 
+## Verify
+
+→ verify: `test -f commit-message/SKILL.md && grep -q '^name:' commit-message/SKILL.md && echo "OK: commit-message valid" || echo "FAIL: commit-message missing"`
+
+
 ## Handoff
 
 Gate: READY -> next: release-branch

package/.pi/skills/craft-skill/SKILL.md

@@ -60,6 +60,10 @@ After drafting, verify:
 - [ ] specs/ output documented if applicable
 - [ ] `sync-skills.sh` run to propagate to Cursor/Gemini
 
+## Verify
+
+→ verify: `test -f craft-skill/SKILL.md && grep -q '^name:' craft-skill/SKILL.md && echo "OK: craft-skill valid" || echo "FAIL: craft-skill missing"`
+
 ---
 
 # Craft Skill — Reference

package/.pi/skills/deepen-architecture/SKILL.md

@@ -93,6 +93,10 @@ Side effects happen inline as decisions crystallize:
 - **User rejects the candidate with a load-bearing reason?** Offer an ADR, framed as: _"Want me to record this as an ADR so future architecture reviews don't re-suggest it?"_ Only offer when the reason would actually be needed by a future explorer. See [ADR-FORMAT.md](../model-domain/ADR-FORMAT.md).
 - **Want to explore alternative interfaces for the deepened module?** See [INTERFACE-DESIGN.md](INTERFACE-DESIGN.md).
 
+## Verify
+
+→ verify: `test -f deepen-architecture/SKILL.md && grep -q '^name:' deepen-architecture/SKILL.md && echo "OK: deepen-architecture valid" || echo "FAIL: deepen-architecture missing"`
+
 ---
 
 # Deepening

package/.pi/skills/define-language/SKILL.md

@@ -79,3 +79,7 @@ When invoked again in the same conversation:
 3. Update definitions if understanding has evolved
 4. Re-flag any new ambiguities
 5. Rewrite the example dialogue to incorporate new terms
+
+## Verify
+
+→ verify: `test -f define-language/SKILL.md && grep -q '^name:' define-language/SKILL.md && echo "OK: define-language valid" || echo "FAIL: define-language missing"`

package/.pi/skills/define-success/SKILL.md

@@ -62,3 +62,7 @@ Add any missing pairs.
 Present the pairs to the user and ask: "Does this capture everything the task requires? Anything missing?"
 
 Once confirmed, these pairs become the skeleton for `plan-work`'s steps. Pass them along when calling `plan-work`.
+
+## Verify
+
+→ verify: `test -f define-success/SKILL.md && grep -q '^name:' define-success/SKILL.md && echo "OK: define-success valid" || echo "FAIL: define-success missing"`

package/.pi/skills/delegate-task/SKILL.md

@@ -76,3 +76,7 @@ Check:
 ```
 
 Report the decision and rationale to the user.
+
+## Verify
+
+→ verify: `test -f delegate-task/SKILL.md && grep -q '^name:' delegate-task/SKILL.md && echo "OK: delegate-task valid" || echo "FAIL: delegate-task missing"`

package/.pi/skills/deploy/SKILL.md

@@ -105,6 +105,10 @@ For comprehensive health-checking, chain to the `smoke-test` skill:
 bash scripts/run-smoke.sh "$DEPLOY_URL"
 ```
 
+## Verify
+
+→ verify: `test -f deploy/SKILL.md && grep -q '^name:' deploy/SKILL.md && echo "OK: deploy valid" || echo "FAIL: deploy missing"`
+
 ---
 
 # Deploy — Reference