@lifeaitools/rdc-skills 0.24.38 → 0.24.41

package/skills/status/SKILL.md

@@ -1,99 +1,99 @@
----
-name: rdc:status
-description: "Usage `rdc:status` — Read-only snapshot of open epics, work items, Coolify health, blockers, and next recommended action. Safe to call anytime. Start here when disoriented or resuming a session."
----
-
-> **⚠️ OUTPUT CONTRACT (READ FIRST):** `guides/output-contract.md`
-> Checklist-only output. No tool-call narration. No raw MCP/JSON/log dumps.
-> One checklist upfront, updated in place, shown again at end with a 1-line verdict.
-
-> If dispatching subagents or running as a subagent: read `{PROJECT_ROOT}/.rdc/guides/agent-bootstrap.md` first (fallback: `{PROJECT_ROOT}/.rdc/guides/agent-bootstrap.md`).
-
-
-# rdc:status — Project Dashboard
-
-## When to Use
-- Start of any session
-- Project lead asks "what's the status", "where are we", "what's open"
-- Before deciding what to work on next
-
-## Procedure
-
-1. **Open epics:**
-   ```sql
-   SELECT get_open_epics();
-   ```
-
-2. **Work item counts by status:**
-   ```sql
-   SELECT status, count(*) FROM work_items 
-   WHERE status != 'archived' 
-   GROUP BY status ORDER BY status;
-   ```
-
-3. **Items by label/project (top 10 labels):**
-   ```sql
-   SELECT unnest(labels) as label, count(*), 
-          count(*) FILTER (WHERE status = 'done') as done,
-          count(*) FILTER (WHERE status = 'todo') as todo,
-          count(*) FILTER (WHERE status = 'in_progress') as wip
-   FROM work_items 
-   GROUP BY label ORDER BY count DESC LIMIT 10;
-   ```
-
-4. **Recent activity (last 48 hours):**
-   ```sql
-   SELECT title, status, updated_at 
-   FROM work_items 
-   WHERE updated_at > now() - interval '48 hours'
-   ORDER BY updated_at DESC LIMIT 15;
-   ```
-
-5. **Git status:**
-   ```bash
-   git log --oneline -10
-   git status
-   git branch -v
-   ```
-
-6. **Dev server health (PM2):**
-   SSH into the Coolify server and check PM2 process state:
-   ```bash
-   # Note: clauth key name for the SSH key may vary by project. Common names: ssh-key-path, vultr-ssh-key, dev-ssh-key.
-   # Check available keys with: curl -s http://127.0.0.1:52437/list-services
-   SSH_KEY=$(curl -s http://127.0.0.1:52437/v/ssh-key-path)
-   ssh -i "$SSH_KEY" -o StrictHostKeyChecking=no root@64.237.54.189 "pm2 jlist 2>/dev/null"
-   ```
-   Parse the JSON array — for each process note `name`, `pm2_env.status` (online/stopped/errored), and `pm2_env.pm_uptime`.
-   If SSH fails or PM2 returns nothing, note "PM2 unreachable" and continue.
-
-7. **Infrastructure health** (if MCP available):
-   - Get infrastructure overview or diagnose issues
-   - Report any apps with failed builds or down containers
-
-8. **Present as a compact dashboard:**
-   ```
-   ## Open Epics (N)
-   <table>
-   
-   ## Work Items: X done | Y todo | Z in_progress
-   
-   ## Recent (48h)
-   <list>
-   
-   ## Dev Servers (PM2)
-   online: studio canvas lifeai ...   stopped: issho-invest
-   
-   ## Deployments
-   <green/red/yellow status>
-   
-   ## Recommended Next
-   <highest priority unstarted epic>
-   ```
-
-## Rules
-- Keep output concise — this is a glance, not a deep dive
-- Always end with a recommendation for what to work on next
-- After the Recommended Next section, suggest which guide file from `.rdc/guides/` (fallback: `.rdc/guides/`) the recommended work would need
-- Use database MCP for queries (not raw curl)
-- If infrastructure MCP is unavailable, skip deployment status and note it
+---
+name: rdc:status
+description: "Usage `rdc:status` — Read-only snapshot of open epics, work items, Coolify health, blockers, and next recommended action. Safe to call anytime. Start here when disoriented or resuming a session."
+---
+
+> **⚠️ OUTPUT CONTRACT (READ FIRST):** `guides/output-contract.md`
+> Checklist-only output. No tool-call narration. No raw MCP/JSON/log dumps.
+> One checklist upfront, updated in place, shown again at end with a 1-line verdict.
+
+> If dispatching subagents or running as a subagent: read `{PROJECT_ROOT}/.rdc/guides/agent-bootstrap.md` first (fallback: `{PROJECT_ROOT}/.rdc/guides/agent-bootstrap.md`).
+
+
+# rdc:status — Project Dashboard
+
+## When to Use
+- Start of any session
+- Project lead asks "what's the status", "where are we", "what's open"
+- Before deciding what to work on next
+
+## Procedure
+
+1. **Open epics:**
+   ```sql
+   SELECT get_open_epics();
+   ```
+
+2. **Work item counts by status:**
+   ```sql
+   SELECT status, count(*) FROM work_items 
+   WHERE status != 'archived' 
+   GROUP BY status ORDER BY status;
+   ```
+
+3. **Items by label/project (top 10 labels):**
+   ```sql
+   SELECT unnest(labels) as label, count(*), 
+          count(*) FILTER (WHERE status = 'done') as done,
+          count(*) FILTER (WHERE status = 'todo') as todo,
+          count(*) FILTER (WHERE status = 'in_progress') as wip
+   FROM work_items 
+   GROUP BY label ORDER BY count DESC LIMIT 10;
+   ```
+
+4. **Recent activity (last 48 hours):**
+   ```sql
+   SELECT title, status, updated_at 
+   FROM work_items 
+   WHERE updated_at > now() - interval '48 hours'
+   ORDER BY updated_at DESC LIMIT 15;
+   ```
+
+5. **Git status:**
+   ```bash
+   git log --oneline -10
+   git status
+   git branch -v
+   ```
+
+6. **Dev server health (PM2):**
+   SSH into the Coolify server and check PM2 process state:
+   ```bash
+   # Note: clauth key name for the SSH key may vary by project. Common names: ssh-key-path, vultr-ssh-key, dev-ssh-key.
+   # Check available keys with: curl -s http://127.0.0.1:52437/list-services
+   SSH_KEY=$(curl -s http://127.0.0.1:52437/v/ssh-key-path)
+   ssh -i "$SSH_KEY" -o StrictHostKeyChecking=no root@64.237.54.189 "pm2 jlist 2>/dev/null"
+   ```
+   Parse the JSON array — for each process note `name`, `pm2_env.status` (online/stopped/errored), and `pm2_env.pm_uptime`.
+   If SSH fails or PM2 returns nothing, note "PM2 unreachable" and continue.
+
+7. **Infrastructure health** (if MCP available):
+   - Get infrastructure overview or diagnose issues
+   - Report any apps with failed builds or down containers
+
+8. **Present as a compact dashboard:**
+   ```
+   ## Open Epics (N)
+   <table>
+   
+   ## Work Items: X done | Y todo | Z in_progress
+   
+   ## Recent (48h)
+   <list>
+   
+   ## Dev Servers (PM2)
+   online: studio canvas lifeai ...   stopped: issho-invest
+   
+   ## Deployments
+   <green/red/yellow status>
+   
+   ## Recommended Next
+   <highest priority unstarted epic>
+   ```
+
+## Rules
+- Keep output concise — this is a glance, not a deep dive
+- Always end with a recommendation for what to work on next
+- After the Recommended Next section, suggest which guide file from `.rdc/guides/` (fallback: `.rdc/guides/`) the recommended work would need
+- Use database MCP for queries (not raw curl)
+- If infrastructure MCP is unavailable, skip deployment status and note it

package/skills/terminal-config/SKILL.md

@@ -1,62 +1,62 @@
----
-name: rdc:terminal-config
-description: "Usage `rdc:terminal-config <task>` — read and safely modify Windows Terminal settings, shell profiles, and agent startup sequencing without relying on machine-specific paths."
----
-
-> **⚠️ OUTPUT CONTRACT (READ FIRST):** `guides/output-contract.md`
-> Checklist-only output. No tool-call narration. No raw MCP/JSON/log dumps.
-> One checklist upfront, updated in place, shown again at end with a 1-line verdict.
-
-# rdc:terminal-config — Terminal & Agent Startup
-
-## When to Use
-
-- Before modifying Windows Terminal settings, shell profiles, keybindings, startup scripts, or agent launch commands.
-- When setting up a Claude, Codex, or project-specific cell profile.
-- When startup is broken because the wrong cwd, shell, env var, or prompt file is being loaded.
-
-## Rules
-
-1. Read the current settings file before editing.
-2. Never change existing profile GUIDs unless the user explicitly wants a new profile identity.
-3. Never remove intentionally null keybindings without explaining the consequence.
-4. Validate JSON before saving.
-5. Use environment-derived paths:
-   - Terminal settings: `$env:LOCALAPPDATA\Packages\Microsoft.WindowsTerminal_8wekyb3d8bbwe\LocalState\settings.json`
-   - Claude user settings: `$env:USERPROFILE\.claude\settings.json`
-   - Codex user config: `$env:USERPROFILE\.codex\config.toml`
-   - Project startup scripts: `{PROJECT_ROOT}\scripts\...`
-
-## Safe Edit Procedure
-
-```
-rdc:terminal-config: <task>
-[ ] Current settings file located from environment
-[ ] File read before edit
-[ ] Existing GUIDs preserved
-[ ] Null keybindings preserved or explicitly approved
-[ ] JSON/TOML syntax validated
-[ ] Startup command uses the intended shell, cwd, and project root
-[ ] Verification command run
-✅ rdc:terminal-config: <result>
-```
-
-## Windows Terminal Notes
-
-- Prefer `pwsh.exe` for PowerShell 7 profiles.
-- Keep `startingDirectory` explicit for project profiles.
-- Use a newly generated GUID for each new profile:
-
-```powershell
-"{$([System.Guid]::NewGuid().ToString())}"
-```
-
-- Validate settings before restarting Terminal:
-
-```powershell
-Get-Content "$env:LOCALAPPDATA\Packages\Microsoft.WindowsTerminal_8wekyb3d8bbwe\LocalState\settings.json" | ConvertFrom-Json
-```
-
-## Startup Scripts
-
-If a project uses role/cell startup scripts, keep those scripts under the project root and commit them with the project. Do not bake one user's absolute machine paths into shared RDC skill files.
+---
+name: rdc:terminal-config
+description: "Usage `rdc:terminal-config <task>` — read and safely modify Windows Terminal settings, shell profiles, and agent startup sequencing without relying on machine-specific paths."
+---
+
+> **⚠️ OUTPUT CONTRACT (READ FIRST):** `guides/output-contract.md`
+> Checklist-only output. No tool-call narration. No raw MCP/JSON/log dumps.
+> One checklist upfront, updated in place, shown again at end with a 1-line verdict.
+
+# rdc:terminal-config — Terminal & Agent Startup
+
+## When to Use
+
+- Before modifying Windows Terminal settings, shell profiles, keybindings, startup scripts, or agent launch commands.
+- When setting up a Claude, Codex, or project-specific cell profile.
+- When startup is broken because the wrong cwd, shell, env var, or prompt file is being loaded.
+
+## Rules
+
+1. Read the current settings file before editing.
+2. Never change existing profile GUIDs unless the user explicitly wants a new profile identity.
+3. Never remove intentionally null keybindings without explaining the consequence.
+4. Validate JSON before saving.
+5. Use environment-derived paths:
+   - Terminal settings: `$env:LOCALAPPDATA\Packages\Microsoft.WindowsTerminal_8wekyb3d8bbwe\LocalState\settings.json`
+   - Claude user settings: `$env:USERPROFILE\.claude\settings.json`
+   - Codex user config: `$env:USERPROFILE\.codex\config.toml`
+   - Project startup scripts: `{PROJECT_ROOT}\scripts\...`
+
+## Safe Edit Procedure
+
+```
+rdc:terminal-config: <task>
+[ ] Current settings file located from environment
+[ ] File read before edit
+[ ] Existing GUIDs preserved
+[ ] Null keybindings preserved or explicitly approved
+[ ] JSON/TOML syntax validated
+[ ] Startup command uses the intended shell, cwd, and project root
+[ ] Verification command run
+✅ rdc:terminal-config: <result>
+```
+
+## Windows Terminal Notes
+
+- Prefer `pwsh.exe` for PowerShell 7 profiles.
+- Keep `startingDirectory` explicit for project profiles.
+- Use a newly generated GUID for each new profile:
+
+```powershell
+"{$([System.Guid]::NewGuid().ToString())}"
+```
+
+- Validate settings before restarting Terminal:
+
+```powershell
+Get-Content "$env:LOCALAPPDATA\Packages\Microsoft.WindowsTerminal_8wekyb3d8bbwe\LocalState\settings.json" | ConvertFrom-Json
+```
+
+## Startup Scripts
+
+If a project uses role/cell startup scripts, keep those scripts under the project root and commit them with the project. Do not bake one user's absolute machine paths into shared RDC skill files.

package/skills/tests/MATRIX.md

@@ -1,54 +1,54 @@
-# RDC Skill Test Suite Matrix
-
-Current coverage: 30 manifests for 30 skill directories.
-
-The manifest layer verifies each skill can be started from a realistic caller prompt in an isolated `RDC_TEST=1` sandbox. The acceptance harness can run either Claude (`--engine claude`) or Codex (`--engine codex`) against the same manifests and records the engine stream, extracted tool calls, stdout/stderr artifacts, rendered assistant output, failures, lessons learned, and next build optimizations under `.rdc/reports/`.
-
-`rdc:channel-formatter` currently has the strongest content acceptance fixture: it asserts source-grounded social-pack output, forbidden-claim absence, and observable corpus or web-search tool routing. `rdc:help` asserts the public MCP/curl caller surface and structured `format:"json"` discovery path. `rdc:self-test` validates the test-tier/evidence language. Planning/reporting skills (`rdc:preplan`, `rdc:plan`, `rdc:handoff`, `rdc:prototype`, `rdc:review`, `rdc:report`) assert their required artifacts and safety boundaries. Operational/content skills (`rdc:brochure`, `rdc:convert`, `lifeai-brochure-author`, `rdc:build`, `rdc:fixit`, `rdc:fs-mcp`, `rdc:workitems`) assert artifact paths, dispatch gates, scoped commits, filesystem boundaries, and sandboxed work-item behavior. Coordination/design/maintenance skills (`rdc:co-develop`, `rdc:collab`, `rdc:design`, `rdc:extract-verifier-rules`, `rdc:housekeeping`, `rdc:overnight`, `rdc:rpms-filemap`) assert relay safety, token/reference use, verifier corpus safety, maintenance/reporting, unattended sentinel behavior, and canonical file-map rules. All manifests now include a deeper acceptance block beyond the basic executable smoke fixture.
-
-| Skill | Manifest | Fixture prompt class | Assertions | Acceptance depth |
-|---|---|---|---|---|
-| `rdc:brochure` | `rdc-brochure.test.json` | HTML-to-PDF brochure fixture | `exit_code`, `stdout_contains` | PDF artifact, page/size/source output, and no upload/deploy/source-modification checks |
-| `rdc:build` | `rdc-build.test.json` | Unattended build from sandbox label | `commits_made`, `exit_code`, `stdout_contains` | Dispatch guide, code-review, validator, sandbox, and no-push/no-skip checks |
-| `rdc:channel-formatter` | `rdc-channel-formatter.test.json` | Long article to social content pack | `exit_code`, `stdout_contains` | Output contains, output not contains, tool-call routing |
-| `rdc:co-develop` | `rdc-co-develop.test.json` | Coordination status | `commits_made`, `exit_code`, `stdout_contains` | codevelop status, sandbox skip, no chitchat/live-send negative checks |
-| `rdc:collab` | `rdc-collab.test.json` | Claude session relay fixture | `commits_made`, `exit_code`, `stdout_contains` | chitchat/SSE session handling, sandbox skip, and no-reply/no-push negative checks |
-| `rdc:convert` | `rdc-convert.test.json` | Markdown-to-Word conversion fixture | `exit_code`, `stdout_contains` | build-corpus command surface, Word/Markdown flags, and no-GUI/global-install/commit checks |
-| `rdc:deploy` | `rdc-deploy.test.json` | Deployment diagnosis | `commits_made`, `exit_code`, `stdout_contains` | Read-only diagnose output and destructive deploy/DNS negative checks |
-| `rdc:edit` | `rdc-edit.test.json` | Local editor launch | `exit_code`, `stdout_contains` | Resolved target, editor host URL, and no-unknown-target negative checks |
-| `rdc:design` | `rdc-design.test.json` | Studio palette audit | `exit_code`, `stdout_contains` | Studio/token/Rampa references and external-write negative checks |
-| `rdc:fixit` | `rdc-fixit.test.json` | Tiny sandbox typo fix | `commits_made`, `exit_code`, `stdout_contains` | Scope, code-review, specific-file staging, and broad-add negative checks |
-| `rdc:fs-mcp` | `rdc-fs-mcp.test.json` | File-system bridge read fixture | `commits_made`, `exit_code`, `stdout_contains` | Local-vs-MCP access boundary, live FS, no-preview/no-overwrite checks, and read-tool routing |
-| `rdc:handoff` | `rdc-handoff.test.json` | Stub work handoff | `exit_code`, `stdout_contains` | Plan path, DoD, work-item handoff, and placeholder negative checks |
-| `rdc:help` | `rdc-help.test.json` | Help menu rendering | `commits_made`, `exit_code`, `stdout_contains` | MCP/curl output, structured JSON fetch, dev-endpoint negative checks |
-| `rdc:housekeeping` | `rdc-housekeeping.test.json` | Read-only housekeeping audit | `commits_made`, `exit_code`, `stdout_contains` | PUBLISH/CLAUDE/orphan/report checks and no-fix/no-Coolify-write negative checks |
-| `lifeai-brochure-author` | `rdc-lifeai-brochure-author.test.json` | JSX compliance review fixture | `exit_code`, `stdout_contains` | brochure-kit component guidance and raw HTML/style/token negative checks |
-| `rdc:overnight` | `rdc-overnight.test.json` | Label-based overnight queue drain | `exit_code`, `stdout_contains` | overnight sentinel, clauth preflight, full chain, and no-wide-add/no-lock-left negative checks |
-| `rdc:plan` | `rdc-plan.test.json` | Health-endpoint planning prompt | `exit_code`, `stdout_contains` | Design-decision, decomposition-matrix, work-package, and coarse-check negative checks |
-| `rdc:preplan` | `rdc-preplan.test.json` | Rate-limiter research prompt | `commits_made`, `exit_code`, `stdout_contains` | Research artifact, comparison, unknowns, recommendation, and research-tool routing checks |
-| `rdc:prototype` | `rdc-prototype.test.json` | Tiny component prototype prompt | `exit_code`, `stdout_contains` | Local TSX prototype artifact, registry, test-mode, and production-write negative checks |
-| `rdc:brochurify` | `rdc-rdc-brochurify.test.json` | Read-only markdown Brochurify fixture | `exit_code`, `stdout_contains` | Sandbox output and no-follow-up negative checks |
-| `rdc:extract-verifier-rules` | `rdc-rdc-extract-verifier-rules.test.json` | Enhancement-log verifier fixture | `exit_code`, `stdout_contains` | Candidate, false-positive, known-good, recall, and no-auto-merge/PR negative checks |
-| `rdc:release` | `rdc-release.test.json` | Dry-run package release | `commits_made`, `exit_code`, `stdout_contains` | Dry-run release checklist and force/bypass negative checks |
-| `rdc:report` | `rdc-report.test.json` | Unattended report generation | `exit_code`, `stdout_contains` | Report status block, counts, next recommendation, and dump/push negative checks |
-| `rdc:review` | `rdc-review.test.json` | Unattended review gate | `exit_code`, `stdout_contains` | Code-review, verify, fresh-evidence, tsc, and forbidden-build negative checks |
-| `rdc:rpms-filemap` | `rdc-rpms-filemap.test.json` | Canonical location lookup | `commits_made`, `exit_code`, `stdout_contains` | Rule #1, context door, one-home rules, and no-hand-edit/second-home negative checks |
-| `rdc:self-test` | `rdc-self-test.test.json` | Strict self-test prompt | `commits_made`, `exit_code`, `stdout_contains` | Strict validation evidence, hook/guide checks, and no-relaxed-linter negative check |
-| `rdc:status` | `rdc-status.test.json` | Read-only status snapshot | `commits_made`, `exit_code`, `stdout_contains` | Read-only status output and raw-MCP/write negative checks |
-| `rdc:terminal-config` | `rdc-terminal-config.test.json` | Hidden-window launch policy audit | `commits_made`, `exit_code`, `stdout_contains` | Hidden-window output and focus/collapse API negative checks |
-| `rdc:watch` | `rdc-watch.test.json` | Watcher initialization prompt | `exit_code`, `stdout_contains` | Test-mode no-open output, tool-call routing, and focus-event negative checks |
-| `rdc:workitems` | `rdc-workitems.test.json` | Work-item epic listing | `commits_made`, `exit_code`, `stdout_contains` | Read-only epics query, sandbox, and work-item write negative checks |
-
-## What The Tests Accomplish
-
-- Verify every installed RDC skill has at least one executable caller-facing prompt.
-- Exercise the `RDC_TEST=1` sandbox contract so build tests do not publish, deploy, or mutate production services.
-- Capture observed engine events and extracted tool calls for periodic inspection.
-- Preserve stdout, stderr, and rendered assistant output as report artifacts.
-- Make content-producing skills eligible for deeper source-fidelity checks through the optional `acceptance` block.
-
-## Next Coverage Upgrades
-
-- Run full scheduled Claude and Codex sweeps regularly, not only changed-skill acceptance, so transcripts and lessons stay fresh.
-- Add richer artifact-content checks for output-producing skills (`rdc:brochure`, `rdc:convert`, `rdc:prototype`, `rdc:report`).
-- Promote recurring transcript review findings into manifest assertions instead of relying on human inspection.
+# RDC Skill Test Suite Matrix
+
+Current coverage: 30 manifests for 30 skill directories.
+
+The manifest layer verifies each skill can be started from a realistic caller prompt in an isolated `RDC_TEST=1` sandbox. The acceptance harness can run either Claude (`--engine claude`) or Codex (`--engine codex`) against the same manifests and records the engine stream, extracted tool calls, stdout/stderr artifacts, rendered assistant output, failures, lessons learned, and next build optimizations under `.rdc/reports/`.
+
+`rdc:channel-formatter` currently has the strongest content acceptance fixture: it asserts source-grounded social-pack output, forbidden-claim absence, and observable corpus or web-search tool routing. `rdc:help` asserts the public MCP/curl caller surface and structured `format:"json"` discovery path. `rdc:self-test` validates the test-tier/evidence language. Planning/reporting skills (`rdc:preplan`, `rdc:plan`, `rdc:handoff`, `rdc:prototype`, `rdc:review`, `rdc:report`) assert their required artifacts and safety boundaries. Operational/content skills (`rdc:brochure`, `rdc:convert`, `lifeai-brochure-author`, `rdc:build`, `rdc:fixit`, `rdc:fs-mcp`, `rdc:workitems`) assert artifact paths, dispatch gates, scoped commits, filesystem boundaries, and sandboxed work-item behavior. Coordination/design/maintenance skills (`rdc:co-develop`, `rdc:collab`, `rdc:design`, `rdc:extract-verifier-rules`, `rdc:housekeeping`, `rdc:overnight`, `rdc:rpms-filemap`) assert relay safety, token/reference use, verifier corpus safety, maintenance/reporting, unattended sentinel behavior, and canonical file-map rules. All manifests now include a deeper acceptance block beyond the basic executable smoke fixture.
+
+| Skill | Manifest | Fixture prompt class | Assertions | Acceptance depth |
+|---|---|---|---|---|
+| `rdc:brochure` | `rdc-brochure.test.json` | HTML-to-PDF brochure fixture | `exit_code`, `stdout_contains` | PDF artifact, page/size/source output, and no upload/deploy/source-modification checks |
+| `rdc:build` | `rdc-build.test.json` | Unattended build from sandbox label | `commits_made`, `exit_code`, `stdout_contains` | Dispatch guide, code-review, validator, sandbox, and no-push/no-skip checks |
+| `rdc:channel-formatter` | `rdc-channel-formatter.test.json` | Long article to social content pack | `exit_code`, `stdout_contains` | Output contains, output not contains, tool-call routing |
+| `rdc:co-develop` | `rdc-co-develop.test.json` | Coordination status | `commits_made`, `exit_code`, `stdout_contains` | codevelop status, sandbox skip, no chitchat/live-send negative checks |
+| `rdc:collab` | `rdc-collab.test.json` | Claude session relay fixture | `commits_made`, `exit_code`, `stdout_contains` | chitchat/SSE session handling, sandbox skip, and no-reply/no-push negative checks |
+| `rdc:convert` | `rdc-convert.test.json` | Markdown-to-Word conversion fixture | `exit_code`, `stdout_contains` | build-corpus command surface, Word/Markdown flags, and no-GUI/global-install/commit checks |
+| `rdc:deploy` | `rdc-deploy.test.json` | Deployment diagnosis | `commits_made`, `exit_code`, `stdout_contains` | Read-only diagnose output and destructive deploy/DNS negative checks |
+| `rdc:edit` | `rdc-edit.test.json` | Local editor launch | `exit_code`, `stdout_contains` | Resolved target, editor host URL, and no-unknown-target negative checks |
+| `rdc:design` | `rdc-design.test.json` | Studio palette audit | `exit_code`, `stdout_contains` | Studio/token/Rampa references and external-write negative checks |
+| `rdc:fixit` | `rdc-fixit.test.json` | Tiny sandbox typo fix | `commits_made`, `exit_code`, `stdout_contains` | Scope, code-review, specific-file staging, and broad-add negative checks |
+| `rdc:fs-mcp` | `rdc-fs-mcp.test.json` | File-system bridge read fixture | `commits_made`, `exit_code`, `stdout_contains` | Local-vs-MCP access boundary, live FS, no-preview/no-overwrite checks, and read-tool routing |
+| `rdc:handoff` | `rdc-handoff.test.json` | Stub work handoff | `exit_code`, `stdout_contains` | Plan path, DoD, work-item handoff, and placeholder negative checks |
+| `rdc:help` | `rdc-help.test.json` | Help menu rendering | `commits_made`, `exit_code`, `stdout_contains` | MCP/curl output, structured JSON fetch, dev-endpoint negative checks |
+| `rdc:housekeeping` | `rdc-housekeeping.test.json` | Read-only housekeeping audit | `commits_made`, `exit_code`, `stdout_contains` | PUBLISH/CLAUDE/orphan/report checks and no-fix/no-Coolify-write negative checks |
+| `lifeai-brochure-author` | `rdc-lifeai-brochure-author.test.json` | JSX compliance review fixture | `exit_code`, `stdout_contains` | brochure-kit component guidance and raw HTML/style/token negative checks |
+| `rdc:overnight` | `rdc-overnight.test.json` | Label-based overnight queue drain | `exit_code`, `stdout_contains` | overnight sentinel, clauth preflight, full chain, and no-wide-add/no-lock-left negative checks |
+| `rdc:plan` | `rdc-plan.test.json` | Health-endpoint planning prompt | `exit_code`, `stdout_contains` | Design-decision, decomposition-matrix, work-package, and coarse-check negative checks |
+| `rdc:preplan` | `rdc-preplan.test.json` | Rate-limiter research prompt | `commits_made`, `exit_code`, `stdout_contains` | Research artifact, comparison, unknowns, recommendation, and research-tool routing checks |
+| `rdc:prototype` | `rdc-prototype.test.json` | Tiny component prototype prompt | `exit_code`, `stdout_contains` | Local TSX prototype artifact, registry, test-mode, and production-write negative checks |
+| `rdc:brochurify` | `rdc-rdc-brochurify.test.json` | Read-only markdown Brochurify fixture | `exit_code`, `stdout_contains` | Sandbox output and no-follow-up negative checks |
+| `rdc:extract-verifier-rules` | `rdc-rdc-extract-verifier-rules.test.json` | Enhancement-log verifier fixture | `exit_code`, `stdout_contains` | Candidate, false-positive, known-good, recall, and no-auto-merge/PR negative checks |
+| `rdc:release` | `rdc-release.test.json` | Dry-run package release | `commits_made`, `exit_code`, `stdout_contains` | Dry-run release checklist and force/bypass negative checks |
+| `rdc:report` | `rdc-report.test.json` | Unattended report generation | `exit_code`, `stdout_contains` | Report status block, counts, next recommendation, and dump/push negative checks |
+| `rdc:review` | `rdc-review.test.json` | Unattended review gate | `exit_code`, `stdout_contains` | Code-review, verify, fresh-evidence, tsc, and forbidden-build negative checks |
+| `rdc:rpms-filemap` | `rdc-rpms-filemap.test.json` | Canonical location lookup | `commits_made`, `exit_code`, `stdout_contains` | Rule #1, context door, one-home rules, and no-hand-edit/second-home negative checks |
+| `rdc:self-test` | `rdc-self-test.test.json` | Strict self-test prompt | `commits_made`, `exit_code`, `stdout_contains` | Strict validation evidence, hook/guide checks, and no-relaxed-linter negative check |
+| `rdc:status` | `rdc-status.test.json` | Read-only status snapshot | `commits_made`, `exit_code`, `stdout_contains` | Read-only status output and raw-MCP/write negative checks |
+| `rdc:terminal-config` | `rdc-terminal-config.test.json` | Hidden-window launch policy audit | `commits_made`, `exit_code`, `stdout_contains` | Hidden-window output and focus/collapse API negative checks |
+| `rdc:watch` | `rdc-watch.test.json` | Watcher initialization prompt | `exit_code`, `stdout_contains` | Test-mode no-open output, tool-call routing, and focus-event negative checks |
+| `rdc:workitems` | `rdc-workitems.test.json` | Work-item epic listing | `commits_made`, `exit_code`, `stdout_contains` | Read-only epics query, sandbox, and work-item write negative checks |
+
+## What The Tests Accomplish
+
+- Verify every installed RDC skill has at least one executable caller-facing prompt.
+- Exercise the `RDC_TEST=1` sandbox contract so build tests do not publish, deploy, or mutate production services.
+- Capture observed engine events and extracted tool calls for periodic inspection.
+- Preserve stdout, stderr, and rendered assistant output as report artifacts.
+- Make content-producing skills eligible for deeper source-fidelity checks through the optional `acceptance` block.
+
+## Next Coverage Upgrades
+
+- Run full scheduled Claude and Codex sweeps regularly, not only changed-skill acceptance, so transcripts and lessons stay fresh.
+- Add richer artifact-content checks for output-producing skills (`rdc:brochure`, `rdc:convert`, `rdc:prototype`, `rdc:report`).
+- Promote recurring transcript review findings into manifest assertions instead of relying on human inspection.

package/skills/tests/README.md

@@ -1,47 +1,47 @@
-# Tier 2 Test Manifests
-
-Each RDC skill directory has a colocated test manifest at `skills/tests/rdc-<name>.test.json`.
-There are currently 30 manifests for 30 skill directories. These drive the Tier 2
-behavioral runner and the build acceptance harness for both Claude and Codex.
-
-See `MATRIX.md` for the current skill-by-skill coverage table.
-
-## Schema
-
-See `../../scripts/lib/manifest-schema.mjs` — the `validateManifest` function
-is the source of truth. Top-level fields:
-
-- `manifest_version` (int) — currently 1
-- `skill` (string) — `rdc:<name>`
-- `description` (string) — what this test verifies
-- `fixture` (object) — prompt, precondition files, env vars
-- `assertions` (object) — what to check after the skill runs
-- `acceptance` (object, optional) — assistant output and recorded tool-call assertions
-- `teardown` (object) — cleanup policy
-
-`acceptance` supports:
-
-- `output_contains` — strings that must appear in the rendered assistant output
-- `output_not_contains` — strings that must not appear in the rendered assistant output
-- `tool_calls_include_any` — at least one recorded tool call with a listed name
-- `tool_calls_include_all` — all listed tool-call names must be recorded
-- `tool_calls_argument_matches` — tool-call argument regex matchers for source/corpus/tool routing checks
-
-## Authoring a manifest
-
-1. Start from an existing manifest in this dir
-2. Set `fixture.env.RDC_TEST = "1"` (required — sandbox contract)
-3. Write assertions tight enough to catch real regressions but not flaky
-4. Run `node scripts/acceptance.mjs --skill rdc:<name>` to capture the Claude transcript, tool calls, and lessons learned
-5. When Codex behavior matters, also run `node scripts/acceptance.mjs --engine codex --skill rdc:<name>` against the same manifest
-6. Commit manifest + any skill changes together
-
-For content-producing skills, include both positive and negative output checks.
-Use `stdout_contains` for expected channel-native structure and source facts;
-use `stdout_not_contains` for source-fidelity violations or explicitly forbidden
-claims.
-
-## Adding a new skill
-
-New skills MUST ship with a manifest. Tier 2 CI will block tag push if a
-new skill lands without one.
+# Tier 2 Test Manifests
+
+Each RDC skill directory has a colocated test manifest at `skills/tests/rdc-<name>.test.json`.
+There are currently 30 manifests for 30 skill directories. These drive the Tier 2
+behavioral runner and the build acceptance harness for both Claude and Codex.
+
+See `MATRIX.md` for the current skill-by-skill coverage table.
+
+## Schema
+
+See `../../scripts/lib/manifest-schema.mjs` — the `validateManifest` function
+is the source of truth. Top-level fields:
+
+- `manifest_version` (int) — currently 1
+- `skill` (string) — `rdc:<name>`
+- `description` (string) — what this test verifies
+- `fixture` (object) — prompt, precondition files, env vars
+- `assertions` (object) — what to check after the skill runs
+- `acceptance` (object, optional) — assistant output and recorded tool-call assertions
+- `teardown` (object) — cleanup policy
+
+`acceptance` supports:
+
+- `output_contains` — strings that must appear in the rendered assistant output
+- `output_not_contains` — strings that must not appear in the rendered assistant output
+- `tool_calls_include_any` — at least one recorded tool call with a listed name
+- `tool_calls_include_all` — all listed tool-call names must be recorded
+- `tool_calls_argument_matches` — tool-call argument regex matchers for source/corpus/tool routing checks
+
+## Authoring a manifest
+
+1. Start from an existing manifest in this dir
+2. Set `fixture.env.RDC_TEST = "1"` (required — sandbox contract)
+3. Write assertions tight enough to catch real regressions but not flaky
+4. Run `node scripts/acceptance.mjs --skill rdc:<name>` to capture the Claude transcript, tool calls, and lessons learned
+5. When Codex behavior matters, also run `node scripts/acceptance.mjs --engine codex --skill rdc:<name>` against the same manifest
+6. Commit manifest + any skill changes together
+
+For content-producing skills, include both positive and negative output checks.
+Use `stdout_contains` for expected channel-native structure and source facts;
+use `stdout_not_contains` for source-fidelity violations or explicitly forbidden
+claims.
+
+## Adding a new skill
+
+New skills MUST ship with a manifest. Tier 2 CI will block tag push if a
+new skill lands without one.

package/skills/tests/rdc-brochure.test.json

@@ -1,34 +1,34 @@
-{
-  "manifest_version": 1,
-  "skill": "rdc:brochure",
-  "description": "Brochure renders a small HTML fixture to a PDF artifact under RDC_TEST without modifying source inputs.",
-  "fixture": {
-    "prompt": "rdc:brochure docs/source/brochure-fixture.html --out .rdc/reports/brochure-fixture.pdf --format Letter",
-    "precondition_files": [
-      {
-        "path": "docs/source/brochure-fixture.html",
-        "content": "<!doctype html><html><head><meta charset=\"utf-8\"><title>Fixture Brochure</title><style>@page{size:Letter;margin:0.75in}body{font-family:serif}h1{color:#234}</style></head><body><h1>Fixture Brochure</h1><p>This is a tiny print-ready brochure fixture for RDC skill acceptance.</p></body></html>"
-      }
-    ],
-    "env": { "RDC_TEST": "1" }
-  },
-  "assertions": {
-    "exit_code": 0,
-    "stdout_contains": ["PDF", "page"]
-  },
-  "acceptance": {
-    "output_contains": [
-      "PDF",
-      "page",
-      "KB",
-      ".rdc/reports/brochure-fixture.pdf"
-    ],
-    "output_not_contains": [
-      "upload",
-      "deploy",
-      "git push",
-      "modify source"
-    ]
-  },
-  "teardown": { "reset_branch": true }
-}
+{
+  "manifest_version": 1,
+  "skill": "rdc:brochure",
+  "description": "Brochure renders a small HTML fixture to a PDF artifact under RDC_TEST without modifying source inputs.",
+  "fixture": {
+    "prompt": "rdc:brochure docs/source/brochure-fixture.html --out .rdc/reports/brochure-fixture.pdf --format Letter",
+    "precondition_files": [
+      {
+        "path": "docs/source/brochure-fixture.html",
+        "content": "<!doctype html><html><head><meta charset=\"utf-8\"><title>Fixture Brochure</title><style>@page{size:Letter;margin:0.75in}body{font-family:serif}h1{color:#234}</style></head><body><h1>Fixture Brochure</h1><p>This is a tiny print-ready brochure fixture for RDC skill acceptance.</p></body></html>"
+      }
+    ],
+    "env": { "RDC_TEST": "1" }
+  },
+  "assertions": {
+    "exit_code": 0,
+    "stdout_contains": ["PDF", "page"]
+  },
+  "acceptance": {
+    "output_contains": [
+      "PDF",
+      "page",
+      "KB",
+      ".rdc/reports/brochure-fixture.pdf"
+    ],
+    "output_not_contains": [
+      "upload",
+      "deploy",
+      "git push",
+      "modify source"
+    ]
+  },
+  "teardown": { "reset_branch": true }
+}