@ludecker/aaac 1.0.0

package/templates/cursor/skills/shared/investigation-lite/SKILL.md

@@ -0,0 +1,38 @@
+---
+name: shared-investigation-lite
+description: >-
+  Lightweight pre-change investigation for create and update. Answers what exists,
+  what depends on it, and what constraints apply. Not user-facing.
+disable-model-invocation: true
+---
+
+# Investigation (lite)
+
+**When:** create and update verb lifecycle — **before** planning.
+
+**Readonly.** Do not edit files.
+
+## Three questions (mandatory)
+
+1. **What exists?** — current files, routes, types, tests for target scope
+2. **What depends on it?** — use [dependency-graph](../dependency-graph/SKILL.md) + domain inventory
+3. **What constraints apply?** — inventory Section 2, policies, object maturity
+
+## Output
+
+```yaml
+exists: [bullets with path evidence]
+depends_on: [domains, modules, objects]
+constraints: [from inventory and policies]
+open_questions: [0–3; empty preferred]
+confidence:
+  architecture: 0.0–1.0
+  requirements: 0.0–1.0
+  scope: 0.0–1.0
+```
+
+Pass output to [validation](../validation/SKILL.md). If any confidence below threshold → **STOP, REQUEST CLARIFICATION**.
+
+## Agents
+
+Reuse readonly specs: [discovery-inventory.md](../../../agents/discovery-inventory.md), [discovery-ssot.md](../../../agents/discovery-ssot.md), [dependency-analysis.md](../../../agents/dependency-analysis.md) — 2–3 parallel max for lite path.

package/templates/cursor/skills/shared/migration/SKILL.md

@@ -0,0 +1,22 @@
+---
+name: shared-migration
+description: >-
+  Schema migration files. Object migration (data layer). Not user-facing.
+disable-model-invocation: true
+---
+
+# Shared migration
+
+## Scope
+
+- `supabase/migrations/` only for DB shape changes
+- One concern per migration; apply via Supabase MCP
+
+## Execution focus
+
+- See [schema](schema/SKILL.md) for RLS and type sync
+- Backward-compatible defaults; document breaking changes
+
+## Must not
+
+- Edit applied production migrations in place — add new migration

package/templates/cursor/skills/shared/model/SKILL.md

@@ -0,0 +1,22 @@
+---
+name: shared-model
+description: >-
+  Entities, DTOs, shared types, query shapes. Object model (data layer). Not user-facing.
+disable-model-invocation: true
+---
+
+# Shared model
+
+## Scope
+
+- `shared/` types, Zod schemas, domain entities in code
+- Align with DB schema; derive — do not duplicate SSOT columns in stores
+
+## Execution focus
+
+- Centralize invariants in schemas
+- Map API ↔ model ↔ persistence at boundaries only
+
+## Fix / check focus
+
+- Nullable vs required, enum coverage, query filters

package/templates/cursor/skills/shared/module-authoring/SKILL.md

@@ -0,0 +1,29 @@
+---
+name: shared-module-authoring
+description: >-
+  Scaffolds domains/<slug>/update/ inventory + orchestrator when slug missing
+  from graph. Discovery swarm required. Not user-facing.
+disable-model-invocation: true
+---
+
+# Shared module-authoring
+
+## When
+
+- `create-feature` for unknown slug
+- Intent `create domain` / greenfield module
+- Graph resolver finds no `domains/<slug>/update/`
+
+## Steps
+
+1. Discovery swarm (4–6 readonly) — inventory + constraints + out-of-scope
+2. Create `domains/<slug>/update/inventory/SKILL.md` from template in [authoring-template.md](authoring-template.md)
+3. Create `domains/<slug>/update/orchestrator/SKILL.md` + `contract.yaml` (copy cms pattern)
+4. Add slug to `graph.yaml` resolvers (`update-module-by-slug`, etc.)
+5. [reporting](../reporting/SKILL.md) — path to new domain
+
+Do not change application code unless a separate code command is active.
+
+## Governance
+
+[implementation](../governance/implementation/SKILL.md) applies only if authoring includes code in same run.

package/templates/cursor/skills/shared/module-authoring/authoring-template.md

@@ -0,0 +1,9 @@
+# Domain inventory template
+
+Copy to `domains/<slug>/update/inventory/SKILL.md` and fill from discovery.
+
+Sections required: Module identity, Constraints, Module inventory (last synced date), Tests, Anti-patterns.
+
+Link `docs/architecture_<slug>.md` when created via `update-doc`.
+
+Orchestrator: copy `domains/cms/update/orchestrator/` and set slug in contract.yaml.

package/templates/cursor/skills/shared/planning/SKILL.md

@@ -0,0 +1,30 @@
+---
+name: shared-planning
+description: >-
+  Plans AAAC changes: SSOT, layers, size budgets, state machines. Readonly until
+  orchestrator approves. Invoked by graph — not user-facing.
+disable-model-invocation: true
+---
+
+# Shared planning
+
+## When
+
+After `discovery_swarm`. **Does not edit files.**
+
+## Agents
+
+Optional parallel readonly:
+
+- [plan-layer-map.md](../../../agents/plan-layer-map.md)
+- [plan-state-machines.md](../../../agents/plan-state-machines.md)
+
+## Plan document (internal)
+
+- User intent → concrete paths (from domain inventory)
+- SSOT owner for new state
+- Extract before add if any file ≥80% budget
+- Named machine states/events if async coordination added
+- Migration + schema steps if DB changes
+
+Orchestrator gates `execute` — no implementation until plan respects domain constraints.

package/templates/cursor/skills/shared/platform-release/SKILL.md

@@ -0,0 +1,46 @@
+---
+name: shared-platform-release
+description: >-
+  Phased release swarm coordination for release-app. Wave 1 git (blocking),
+  Wave 2 Render deploy. Internal only.
+disable-model-invocation: true
+---
+
+# Shared platform release
+
+## Swarm DAG
+
+```text
+Preflight (optional tests)
+  ↓
+Wave 1: release-git  ← BLOCKING
+  ↓
+Wave 2: release-render
+  ↓
+Parent: verification + reporting
+```
+
+## Wave 1 — launch one agent
+
+```
+Task (shell or generalPurpose): release-git
+Read: agents/release-git.md, ship-procedure.md
+Return: commit_sha, commit_message_*
+On failure: abort pipeline
+```
+
+## Wave 2 — Render deploy
+
+Only after Wave 1 `status: success`. Pass `commit_sha` and messages.
+
+| Agent spec | Tool |
+|------------|------|
+| [release-render.md](../../../agents/release-render.md) | generalPurpose or render-assistant MCP |
+
+## Preflight
+
+If intent contains "test" / "with tests": run `pnpm typecheck` before Wave 1.
+
+## Reference
+
+Full step detail: [ship-procedure.md](ship-procedure.md)

package/templates/cursor/skills/shared/platform-release/orchestrator/SKILL.md

@@ -0,0 +1,51 @@
+---
+name: release-app-orchestrator
+description: >-
+  Orchestrates release-app with phased expert subagents: git (blocking),
+  then Render deploy verification.
+disable-model-invocation: true
+---
+
+# release-app orchestrator
+
+## Parse
+
+- **Domain:** `production` (default) or environment token
+- **Intent:** quoted ship notes (optional) — e.g. `"Ship with tests"`
+
+## Load
+
+1. [../SKILL.md](../SKILL.md) — swarm DAG
+2. [ship-procedure.md](../ship-procedure.md) — step reference
+3. [graph.yaml](../../../../aaac/graph.yaml) — `release-app`
+4. [ludecker-infrastructure](../../../ludecker/infrastructure/SKILL.md)
+
+## Phases
+
+### 0. Preflight
+
+If intent requests tests: run `pnpm typecheck` before any git work ([shared/testing](../../testing/SKILL.md) patterns).
+
+### 1. Wave 1 — Git (blocking)
+
+Spawn subagent per [agents/release-git.md](../../../../agents/release-git.md).
+
+**Do not** start Wave 2 until `commit_sha` is returned.
+
+### 2. Wave 2 — Render
+
+Spawn [release-render](../../../../agents/release-render.md) with `commit_sha`, `commit_message_first_line`, `commit_message_body`.
+
+If Render `build_failed`, overall status is failed.
+
+### 3. Verify + report
+
+- [verification](../../verification/SKILL.md): deploy live, intent met
+- [reporting](../../reporting/SKILL.md): layman summary + technical details table per agent
+
+## Anti-patterns
+
+- Starting Render before push completes
+- Single agent doing git + deploy without expert prompts
+- Using `plugin-render-render` MCP
+- Force-pushing `main`

package/templates/cursor/skills/shared/platform-release/orchestrator/contract.yaml

@@ -0,0 +1,29 @@
+name: release-app-orchestrator
+purpose: Ship to main with phased expert subagents (git, then parallel integrations)
+inputs:
+  domain:
+    required: false
+    default: production
+  intent:
+    required: false
+outputs:
+  commit_sha:
+    type: string
+  deploy_status:
+    type: string
+  report:
+    type: markdown
+success_criteria:
+  - git push to main succeeded
+  - render deploy reached live or build failure reported with logs
+  - posthog and linear agents run after git (parallel wave)
+failure_conditions:
+  - git push failed
+  - wave 2 started before commit_sha known
+dependencies:
+  skills: [platform-release, testing, verification, reporting]
+  policies: [master-rules, mcp-and-deploy]
+verification:
+  - git_pushed
+  - render_polled
+  - agent_wave_order_respected

package/templates/cursor/skills/shared/platform-release/ship-procedure.md

@@ -0,0 +1,31 @@
+# Release ship procedure (reference)
+
+Canonical steps migrated from legacy `/ship-ludecker`. Subagents in `agents/release-*.md` own each slice.
+
+## Git (Wave 1 — blocking)
+
+1. Confirm repo: `signalbynoise/ludecker` (or local ludecker monorepo)
+2. `git status` + `git diff` + `git diff --staged` + `git log -5 --oneline`
+3. Never stage `.env`, `.env.local`, credentials, or API keys
+4. Draft 1–2 sentence commit message from diff and user intent
+5. `git add` intentional paths → `git commit` (HEREDOC message)
+6. Ensure on `main`: `git checkout main` if needed; `git pull --rebase origin main`
+7. `git push origin main` — on reject, rebase once more; never force-push main
+8. Output: `commit_sha`, `commit_message_first_line`, `commit_message_body`
+
+**Rules:** No force-push main. On pre-commit hook failure: fix and new commit — never amend unless user asked.
+
+## Render (Wave 2 — after push)
+
+MCP: `user-render` only (not `plugin-render-render`).
+
+**Service SSOT:** `ludecker-website` (see `render.yaml`, `docs/deployment.md`).
+
+1. `list_services` — find `name === "ludecker-website"`, note `id`
+2. After push: `list_deploys` (`limit: 5`) — match deploy `commit.id` to `commit_sha`
+3. Poll up to **15 minutes**, every **30s**, until `status === "live"` or terminal failure
+4. Smoke check: `curl -fsS -o /dev/null -w "%{http_code}" https://ludecker-website.onrender.com/` — expect **200**
+
+## Preflight (optional, Wave 0)
+
+If intent includes "with tests": run `pnpm typecheck` from repo root before git work.

package/templates/cursor/skills/shared/remove/SKILL.md

@@ -0,0 +1,28 @@
+---
+name: shared-remove
+description: >-
+  Delete or retire code, routes, flags, migrations (with caution). Not user-facing.
+disable-model-invocation: true
+---
+
+# Shared remove
+
+## When
+
+`remove-*` commands after discovery confirms scope.
+
+## Mandatory before edits
+
+1. List blast radius (imports, routes, tests, docs, feature flags)
+2. Confirm no production dependency (grep + Fallow if available)
+3. Prefer feature-flag off before hard delete when rollout risk exists
+
+## Execution focus
+
+- Remove dead code paths; update inventory/docs in same change
+- Schema: deprecate column before drop when data may exist
+- Never remove secrets or env from blueprint without render.yaml update
+
+## Must not
+
+- Remove `app` shell or auth entry without explicit user confirmation in intent

package/templates/cursor/skills/shared/reporting/SKILL.md

@@ -0,0 +1,43 @@
+---
+name: shared-reporting
+description: >-
+  AAAC user-facing report template. Plain language first. Not user-facing as a command.
+disable-model-invocation: true
+---
+
+# Shared reporting
+
+## Template
+
+```markdown
+## [Command] — [Short title]
+
+**Bottom line:** [One sentence]
+
+### What we did
+- …
+
+### Verification
+- Tests: …
+- Intent: …
+- Confidence: architecture / requirements / scope
+- Fitness: pass/warning/fail summary
+- Impact: blast_radius, affected domains
+
+### Rollback (if applicable)
+- Plan reference or "not required"
+
+### Skill synced
+- [If applicable] domains/<slug>/update/inventory refreshed
+
+### Follow-ups
+- …
+
+---
+<details>
+<summary>Technical details</summary>
+…
+</details>
+```
+
+Layman sections: no unexplained jargon. Put paths and commit refs in Technical details only.

package/templates/cursor/skills/shared/rollback/SKILL.md

@@ -0,0 +1,46 @@
+---
+name: shared-rollback
+description: >-
+  Rollback plan before execute — files, migrations, deploy. Required for protected
+  maturity and high blast_radius. Not user-facing.
+disable-model-invocation: true
+---
+
+# Rollback planning
+
+**When:** Verb lifecycle phase `rollback` — before execute.
+
+## Question
+
+```text
+If this fails, how do we undo it?
+```
+
+## Required when
+
+- Object maturity **protected** (schema, migration, architecture)
+- `impact_analysis.blast_radius: high`
+- Database, auth, API, or deployment touch in plan
+
+## Output (mandatory)
+
+```yaml
+rollback:
+  files:
+    - path/to/revert or "git checkout -- <paths>"
+  migrations:
+    - reverse_sql: "description or down migration path"
+    - note: "apply via Supabase MCP only with user confirm"
+  deployments:
+    - previous_release: "revert commit sha or Render rollback procedure"
+  data:
+    - backup_needed: true | false
+    - notes: optional
+verified: true | false
+```
+
+If rollback cannot be defined for a protected change → **STOP, REQUEST CLARIFICATION**.
+
+## Execute phase
+
+Execution skill references this plan; does not invent rollback during execute.

package/templates/cursor/skills/shared/root-cause/SKILL.md

@@ -0,0 +1,24 @@
+---
+name: shared-root-cause
+description: >-
+  Deep root-cause framing after investigation on fix paths. Not user-facing.
+disable-model-invocation: true
+---
+
+# Root cause (fix only)
+
+**When:** fix verb lifecycle — after [investigation](../investigation/SKILL.md), before planning.
+
+## Output (mandatory)
+
+```yaml
+symptom: one line
+root_cause: hypothesis with evidence (path:line)
+contributing_factors: [optional bullets]
+fix_strategy: minimal correct change (not symptom patch)
+regression_risk: low | medium | high
+```
+
+If root cause confidence < 0.7 → **STOP, REQUEST CLARIFICATION** — do not plan or execute.
+
+Feed `fix_strategy` and `regression_risk` into [impact-analysis](../impact-analysis/SKILL.md) and [rollback](../rollback/SKILL.md).

package/templates/cursor/skills/shared/run/SKILL.md

@@ -0,0 +1,64 @@
+---
+name: shared-run
+description: >-
+  Create, update, and resume AAAC Runs. All execution state and observability
+  lives on the Run manifest. Not user-facing.
+disable-model-invocation: true
+---
+
+# Run management
+
+**SSOT:** [run/schema.json](../../../aaac/run/schema.json), [run/RUN.md](../../../aaac/run/RUN.md)
+
+Every AAAC command executes within a Run. No standalone execution-state or markdown logs.
+
+## Create Run
+
+At dispatch step 2.5:
+
+1. `run_id` = `run_{YYYYMMDD}_{slug}` from command + object/domain
+2. Load `verb_runtime.<verb>` from graph → initial `pending`
+3. Write `state/runs/{run_id}/run.json`
+4. Set `status: running`, first phase, `phase_kind` from phases.json (`gate: true` → gate)
+
+## Update after each phase
+
+1. Append `log[]` entry (`phase_start` / `phase_complete` / `gate_pass` / `gate_fail`)
+2. Move phase from `pending` to `completed`
+3. Write `checkpoints/{phase}.json`
+4. Store phase outputs in `artifacts/` and reference in `artifacts{}`
+5. Append routing choices to `decisions[]`
+6. Update `phase`, `updated_at`; persist manifest
+
+## Gate failure → human approval
+
+```yaml
+status: blocked
+awaiting_approval: true
+blocked_reason: "<specific reason>"
+phase_kind: gate
+```
+
+STOP. On user approval: log decision, clear `awaiting_approval`, set `status: running`, retry gate.
+
+## Capability resolution (record on Run)
+
+When resolving object capabilities:
+
+```yaml
+capabilities_resolved:
+  layer-boundaries:
+    providers:
+      - id: architecture
+        type: skill
+```
+
+Include MCP providers (`type: mcp`) in decisions even though graph skills exclude them.
+
+## Resume
+
+Read `state/runs/{run_id}/run.json`. Continue from `phase` when `status` is `running` or user approved `blocked`.
+
+## Complete
+
+After `report`: set `status: completed`, write `artifacts.report`.

package/templates/cursor/skills/shared/schema/SKILL.md

@@ -0,0 +1,24 @@
+---
+name: shared-schema
+description: >-
+  Data models, migrations, RLS. Used when object is schema. Not user-facing.
+disable-model-invocation: true
+---
+
+# Shared schema
+
+## Scope
+
+- `supabase/migrations/`, shared Zod/TS types, RLS policies
+- SSOT: schema lives in migrations + shared types — never duplicate columns in app state
+
+## Execution focus
+
+- New migration file per change; apply via Supabase MCP (`hjadkzfemzuvhpwbixbt`)
+- Backward-compatible defaults; document breaking changes in report
+- RLS and security advisors after apply
+
+## Check / test focus
+
+- Migration order, nullable vs required, FK integrity
+- Types consumed by server and app match migration

package/templates/cursor/skills/shared/testing/SKILL.md

@@ -0,0 +1,24 @@
+---
+name: shared-testing
+description: >-
+  Runs vitest and Fallow check_changed for AAAC workflows. Software pass/fail —
+  not goal verification. Not user-facing.
+disable-model-invocation: true
+---
+
+# Shared testing
+
+## When
+
+Phase `test` (and `test_only` orchestrators).
+
+## Steps
+
+1. Run tests from domain inventory relevant to change
+2. Invoke [unit-test-run.md](../../../agents/unit-test-run.md) pattern for targeted vitest
+3. Fallow MCP → `check_changed` on touched files when configured
+4. `ReadLints` on edited paths
+
+## Output
+
+Pass/fail summary with test names and Fallow verdict for `verification` skill.

package/templates/cursor/skills/shared/testing/orchestrator/SKILL.md

@@ -0,0 +1,22 @@
+---
+name: test-function-orchestrator
+description: Orchestrates test-function — targeted journey/integration verification. Internal only.
+disable-model-invocation: true
+---
+
+# test-function orchestrator
+
+## Parse
+
+- **Intent:** required — describes user journey (e.g. "add new user to project")
+- **Domain:** optional slug in args for scope hint
+
+## Phases
+
+1. [discovery](../discovery/SKILL.md) — map journey to files/tests
+2. [planning](../planning/SKILL.md) — which tests to run or add
+3. [testing](../testing/SKILL.md) — run vitest/e2e as appropriate
+4. [verification](../verification/SKILL.md) — journey criteria met?
+5. [reporting](../reporting/SKILL.md)
+
+Code changes only if intent explicitly requests fixing failures (then `execution`).

package/templates/cursor/skills/shared/validation/SKILL.md

@@ -0,0 +1,56 @@
+---
+name: shared-validation
+description: >-
+  Confidence gates before execute. STOP and request clarification when thresholds
+  not met. Not user-facing.
+disable-model-invocation: true
+---
+
+# Validation (confidence gates)
+
+**When:** After plan, **before** impact_analysis / execute.
+
+## Thresholds (SSOT)
+
+From [ontology.json](../../../aaac/ontology.json) `confidence`:
+
+| Dimension | Minimum |
+|-----------|---------|
+| architecture | 0.9 |
+| requirements | 0.8 |
+| scope | 0.8 |
+
+## Inputs
+
+- Plan from [planning](../planning/SKILL.md)
+- Confidence scores from [investigation-lite](../investigation-lite/SKILL.md) or [investigation](../investigation/SKILL.md) + [root-cause](../root-cause/SKILL.md)
+- Domain inventory constraints
+
+## Procedure
+
+1. Score each dimension 0.0–1.0 with one-line evidence
+2. Compare to thresholds
+3. If **any** below threshold:
+
+```yaml
+status: blocked
+awaiting_approval: true
+blocked_reason: "confidence.{dimension} {score} below {threshold}"
+```
+
+```text
+STOP — awaiting approval
+Reason: {blocked_reason}
+Run: {run_id}
+```
+
+List specific questions for the user. **Do not proceed to execute** until user approves in chat.
+
+4. Record scores on Run `confidence` and gate result in `gates.results.validate`
+5. If at threshold: emit gate pass, continue gate stack
+
+## Plan sanity checks
+
+- Plan respects inventory out-of-scope
+- Plan names files to touch (no vague "update CMS")
+- Protected/critical objects include rollback mention in plan or next rollback phase