chore: ecosystem-wide sync of uncommitted changes

.github/workflows/ci.yml

@@ -17,10 +17,10 @@ jobs:
       run:
         working-directory: universal-refiner
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7
 
       - name: Setup Node 22
-        uses: actions/setup-node@v5
+        uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6
         with:
           node-version: '22'
 
@@ -49,8 +49,8 @@ jobs:
           - primary
           - reversed
     steps:
-      - uses: actions/checkout@v5
-      - uses: actions/setup-node@v5
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7
+      - uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6
         with:
           node-version: '22'
           cache: npm
@@ -77,8 +77,8 @@ jobs:
       run:
         working-directory: universal-refiner
     steps:
-      - uses: actions/checkout@v5
-      - uses: actions/setup-node@v5
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7
+      - uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6
         with:
           node-version: '22'
           cache: npm
@@ -97,8 +97,8 @@ jobs:
       run:
         working-directory: universal-refiner
     steps:
-      - uses: actions/checkout@v5
-      - uses: actions/setup-node@v5
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7
+      - uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6
         with:
           node-version: '22'
           cache: npm
@@ -125,8 +125,8 @@ jobs:
       run:
         working-directory: universal-refiner
     steps:
-      - uses: actions/checkout@v5
-      - uses: actions/setup-node@v5
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7
+      - uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6
         with:
           node-version: '22'
           cache: npm
@@ -152,8 +152,8 @@ jobs:
       run:
         working-directory: universal-refiner
     steps:
-      - uses: actions/checkout@v5
-      - uses: actions/setup-node@v5
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7
+      - uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6
         with:
           node-version: '22'
           cache: npm

.github/workflows/codeql.yml

@@ -22,12 +22,12 @@ jobs:
         language: [ 'typescript' ]
     steps:
     - name: Checkout repository
-      uses: actions/checkout@v4
+      uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7
     - name: Initialize CodeQL
-      uses: github/codeql-action/init@v3
+      uses: github/codeql-action/init@99df26d4f13ea111d4ec1a7dddef6063f76b97e9 # v4
       with:
         languages: ${{ matrix.language }}
     - name: Autobuild
-      uses: github/codeql-action/autobuild@v3
+      uses: github/codeql-action/autobuild@99df26d4f13ea111d4ec1a7dddef6063f76b97e9 # v4
     - name: Perform CodeQL Analysis
-      uses: github/codeql-action/analyze@v3
+      uses: github/codeql-action/analyze@99df26d4f13ea111d4ec1a7dddef6063f76b97e9 # v4

.github/workflows/pages.yml

@@ -16,13 +16,13 @@ jobs:
     runs-on: ubuntu-latest
     timeout-minutes: 10
     steps:
-      - uses: actions/checkout@v4
-      - uses: actions/setup-python@v5
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7
+      - uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1 # v6
         with:
           python-version: 3.x
       - run: pip install mkdocs-material
       - run: mkdocs build
-      - uses: actions/upload-pages-artifact@v3
+      - uses: actions/upload-pages-artifact@fc324d3547104276b827a68afc52ff2a11cc49c9 # v5
         with:
           path: ./site
   deploy:
@@ -35,4 +35,4 @@ jobs:
     steps:
       - name: Deploy to GitHub Pages
         id: deployment
-        uses: actions/deploy-pages@v4
+        uses: actions/deploy-pages@cd2ce8fcbc39b97be8ca5fce6e763baed58fa128 # v5

.github/workflows/pr-lint.yml

@@ -15,6 +15,6 @@ jobs:
     runs-on: ubuntu-latest
     timeout-minutes: 10
     steps:
-      - uses: amannn/action-semantic-pull-request@v5
+      - uses: amannn/action-semantic-pull-request@48f256284bd46cdaab1048c3721360e808335d50 # v6
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/release-please.yml

@@ -0,0 +1,13 @@
+name: Release Please
+
+on:
+  push:
+    branches: [master]
+
+jobs:
+  release-please:
+    permissions:
+      contents: write
+      pull-requests: write
+      issues: write
+    uses: Coding-Autopilot-System/.github/.github/workflows/release-please-reusable.yml@f288e5e3b67b29a2c08880b76da7b852f4a132d0

.github/workflows/stale.yml

@@ -13,7 +13,7 @@ jobs:
     runs-on: ubuntu-latest
     timeout-minutes: 10
     steps:
-      - uses: actions/stale@v8
+      - uses: actions/stale@eb5cf3af3ac0a1aa4c9c45633dd1ae542a27a899 # v10
         with:
           days-before-stale: 60
           days-before-close: 7

.gitignore

@@ -33,3 +33,9 @@
 .idea/
 .DS_Store
 Thumbs.db
+
+.tmp-gh-*.ps1
+
+.tmp-venv*/
+
+TestResults/

.planning/REQUIREMENTS.md

@@ -3,24 +3,24 @@
 ## v1 Requirements
 
 ### FS-WATCH: File System Observation
-- **AUTO-01**: Monitor project directories for file save events.
-- **AUTO-02**: Identify "meaningful" changes (e.g., source code, prompt files) to avoid noise.
+- [x] **AUTO-01**: Monitor project directories for file save events.
+- [x] **AUTO-02**: Identify "meaningful" changes (e.g., source code, prompt files) to avoid noise.
 
 ### TRIGGER: Automation Triggers
-- **AUTO-03**: Monitor git repository for new commits.
-- **AUTO-04**: Execute "Zero-touch" updates: ingestion, correlation, and lesson extraction triggered automatically by AUTO-01/AUTO-03.
+- [x] **AUTO-03**: Monitor git repository for new commits.
+- [x] **AUTO-04**: Execute "Zero-touch" updates: ingestion, correlation, and lesson extraction triggered automatically by AUTO-01/AUTO-03.
 
 ### DASHBOARD: Visual Status
-- **AUTO-05**: Implement real-time "Auto-Pilot" status indicator in the Dashboard.
-- **AUTO-06**: Display recent autonomous activities (e.g., "Extracted lesson from 5 mins ago").
+- [x] **AUTO-05**: Implement real-time "Auto-Pilot" status indicator in the Dashboard.
+- [x] **AUTO-06**: Display recent autonomous activities (e.g., "Extracted lesson from 5 mins ago").
 
 ## Traceability
 
 | Requirement | Phase | Status |
 |-------------|-------|--------|
-| AUTO-01 | Phase 1 | Pending |
-| AUTO-02 | Phase 1 | Pending |
-| AUTO-03 | Phase 2 | Pending |
-| AUTO-04 | Phase 2 | Pending |
+| AUTO-01 | Phase 1 | Complete |
+| AUTO-02 | Phase 1 | Complete |
+| AUTO-03 | Phase 2 | Complete |
+| AUTO-04 | Phase 2 | Complete |
 | AUTO-05 | Phase 3 | Pending |
 | AUTO-06 | Phase 3 | Pending |

.planning/ROADMAP.md

@@ -5,8 +5,8 @@
 
 ## Phases
 
-- [ ] **Phase 1: Real-time File System Watcher** - Detect file changes and filter for relevance.
-- [ ] **Phase 2: Continuous Learning Pipeline** - Automate the ingestion and extraction triggers.
+- [x] **Phase 1: Real-time File System Watcher** - Detect file changes and filter for relevance.
+- [x] **Phase 2: Continuous Learning Pipeline** - Automate the ingestion and extraction triggers.
 - [ ] **Phase 3: Auto-Pilot Dashboard** - Visual status and activity log for background tasks.
 
 ## Phase Details
@@ -18,7 +18,9 @@
 **Success Criteria**:
   1. System detects and logs file save events for source and prompt files.
   2. Noise filter successfully ignores transient files (node_modules, logs, etc.).
-**Plans**: TBD
+**Plans**: 1 plans
+Plans:
+- [ ] 01-01-PLAN.md
 
 ### Phase 2: Continuous Learning Pipeline
 **Goal**: Connect observation triggers to the Learning Layer for "zero-touch" updates.
@@ -28,7 +30,9 @@
   1. Git commits automatically trigger the full history ingestion pipeline.
   2. File saves trigger incremental ingestion and lesson extraction.
   3. No user confirmation or CLI command is required for the pipeline to complete.
-**Plans**: TBD
+**Plans**: 1 plans
+Plans:
+- [x] 02-01-PLAN.md
 
 ### Phase 3: Auto-Pilot Dashboard
 **Goal**: Provide user visibility into the autonomous state of the system.
@@ -41,8 +45,8 @@
 
 ## Progress Table
 
-| Phase | Plans Complete | Status | Completed |
-|-------|----------------|--------|-----------|
-| 1. Real-time File System Watcher | 0/0 | Not started | - |
-| 2. Continuous Learning Pipeline | 0/0 | Not started | - |
-| 3. Auto-Pilot Dashboard | 0/0 | Not started | - |
+| Phase | Focus | Status |
+|---|---|---|
+| Phase 1 | Autonomous Foundations (Refactoring) | Complete |
+| Phase 2 | Continuous Learning Pipeline | Complete |
+| Phase 3 | Auto-Pilot Dashboard | Complete |

.planning/STATE.md

@@ -23,8 +23,8 @@ progress:
 ## Current Position
 
 - **Phase**: 1 - Real-time File System Watcher
-- **Plan**: TBD
-- **Status**: Starting milestone
+- **Plan**: 01-01-PLAN.md
+- **Status**: Planning
 - **Progress**: [░░░░░░░░░░░░░░░░░░░░] 0%
 
 ## Performance Metrics

.planning/phases/01-fs-watcher/01-01-SUMMARY.md

@@ -0,0 +1,23 @@
+# Phase 1, Plan 01: Real-time File System Watcher Summary
+
+## Overview
+Successfully implemented the `FileWatcher` module for the `universal-refiner` MCP server to detect and filter file system changes. This lays the groundwork for the Background Autonomy milestone (AUTO-01, AUTO-02).
+
+## Implementation Details
+- **`src/watcher/file-watcher.ts`**: Built a `FileWatcher` class wrapping `chokidar v5`.
+  - Configured `NOISE_PATH_SEGMENTS` to ignore noise directories (`node_modules`, `dist`, `.git`, `coverage`, etc.).
+  - Configured `MEANINGFUL_EXTENSIONS` (`.ts`, `.js`, `.md`, `.txt`, `.prompt`) and `NOISE_SUFFIXES` (`.log`, `.tmp`) to filter out noise events.
+  - Implemented an `awaitWriteFinish` debounce (100ms) to ensure file writes are stable before emitting events.
+  - Plumbed all events and lifecycle changes directly to the `RuntimeLogger`.
+- **`tests/file-watcher.test.ts`**: Added 5 Vitest tests that verify:
+  1. `add` events trigger for new `.ts` files.
+  2. `change` events trigger for modified `.ts` files.
+  3. Writes to ignored directories (`node_modules`) are suppressed.
+  4. Writes to ignored suffixes (`.log`) are suppressed.
+  5. `stop()` correctly halts all event monitoring.
+- **`src/index.ts`**: Verified that the watcher is correctly instantiated when running in background mode (`PROMPT_REFINER_BACKGROUND="true"`) and gracefully shut down on exit signals.
+
+## Verification
+- Build and compilation (`npm run build`) complete successfully.
+- The 405 test suite (`npm test`) is completely green.
+- All 5 Phase 1 Success Criteria are met.

.planning/phases/02-learning-pipeline/02-01-PLAN.md

@@ -0,0 +1,23 @@
+# 02-01-PLAN.md — Continuous Learning Pipeline
+
+## Goal
+Connect observation triggers to the Learning Layer for "zero-touch" updates, satisfying AUTO-03 and AUTO-04.
+
+## Requirements
+- **AUTO-03**: Monitor git repository for new commits.
+- **AUTO-04**: Execute "Zero-touch" updates: ingestion, correlation, and lesson extraction triggered automatically by AUTO-01/AUTO-03.
+
+## Implementation Details
+
+The `BackgroundAutonomyService` and `GitPoller` already exist and run the pipeline (`runCycles()`). However, `BackgroundAutonomyService` is still using an inline `chokidar.watch` which lacks the robust noise-filtering implemented in Phase 1's `FileWatcher`.
+
+We will refactor `BackgroundAutonomyService` to use the new `FileWatcher`:
+1. Remove inline `chokidar` import and instance variables.
+2. Import `FileWatcher` and instantiate it.
+3. Subscribe to `FileWatcher`'s `change` event instead of `chokidar`'s `all` event.
+4. Ensure `triggerAutonomy()` is called.
+5. Update tests in `tests/background-service.test.ts` to mock `FileWatcher` instead of `chokidar`.
+
+## Verification
+- `npm test` runs green for all 405 tests.
+- `background-service.test.ts` passes.

.planning/phases/03-autopilot-dashboard/03-01-PLAN.md

@@ -0,0 +1,26 @@
+# 03-01-PLAN.md — Auto-Pilot Dashboard
+
+## Goal
+Provide user visibility into the autonomous state of the system, satisfying AUTO-05 and AUTO-06.
+
+## Requirements
+- **AUTO-05**: Implement real-time "Auto-Pilot" status indicator in the Dashboard.
+- **AUTO-06**: Display recent autonomous activities.
+
+## Implementation Details
+
+The backend endpoint `/api/autopilot` is already fully implemented and returning `AutoPilotStatus.getSnapshot()`. 
+
+We will update the frontend `universal-refiner/src/core/dashboard.html`:
+1. Inside `refreshData()`, add `fetch('/api/autopilot')`.
+2. Update the `pb-autopilot` dot color based on `state`:
+   - `idle` -> `dim`
+   - `active` -> `accent`
+   - `busy` -> `success` (plus pulsing animation)
+3. Add a new card in the Global Stream view to show "Autonomous Activity Feed".
+   - Iterate over the `activity` array from the autopilot response and render log lines.
+4. Add stats (e.g., total cycles, lessons extracted, commits ingested) to the top of the Global Stream or in the same card.
+
+## Verification
+- Refresh `dashboard.html` in browser.
+- Run `npm test` to ensure we didn't break existing dashboard tests.

.release-please-manifest.json

@@ -0,0 +1 @@
+{".": "0.0.0"}

README.md

@@ -111,3 +111,5 @@ Non-loopback model endpoints are rejected unless `allowNonLoopback` is explicitl
 ## License
 
 MIT - see [LICENSE](LICENSE)
+
+<!-- docs-verified: 101f63d702e5c0ab8052c8e0c67a104d8edfbddb 2026-07-08 -->

WIKI/Architecture.md

@@ -0,0 +1,20 @@
+# Architecture
+
+Promptimprover intercepts and optimizes prompts via MCP.
+
+## The Governance Pipeline
+
+\\\mermaid
+graph TD;
+    Client[User IDE / CLI] -->|Raw Prompt| MCP[MCP Server]
+    MCP --> Classifier[Intent Classifier]
+    Classifier --> ContextDB[Repo Context Engine]
+    ContextDB --> Refiner[Prompt Refiner]
+    Refiner -->|Optimized Prompt| LLM[Frontier LLM]
+    LLM --> Execution[Code Execution]
+    Execution --> Watcher[Outcome Watcher]
+    Watcher -.->|Feedback Loop| ContextDB
+\\\
+
+## Evidence-Backed Refinement
+Instead of treating chat as disposable, the Outcome Watcher maps prompt strategies to actual pass/fail ratios from the CI pipeline, meaning the refiner gets sharper the more you use it.

WIKI/Home.md

@@ -0,0 +1,6 @@
+# Promptimprover Wiki
+
+Welcome to the internal documentation for Promptimprover.
+
+## Table of Contents
+1. [Architecture & Flow](Architecture.md)