refactor: restructure into proper project layout + improve docs

Project layout:
- src/: application modules (run_server, auto_merge_bot, health-check, sync-key, trivial_merge, callback_server, start_server)
- scripts/: setup/deployment helpers (setup_all, setup_app, generate_manifest)
- templates/: manifest.json (GitHub App manifest template)
- docs/  + CONTRIBUTING.md: documentation

Improvements:
- flake.nix: added pr-agent-auto-merge wrapper binary, updated installPhase paths
- deploy.yml: syntax check covers all modules including health-check.py and sync-key.py
- README.md: comprehensive with architecture, layout, dev, ops, deployment
- CONTRIBUTING.md: standards and testing checklist
- .gitignore: added *.log, *.pid, .env.*
- Cleanup: removed duplicate manifest_current.json / manifest_final.json
- Fix: health-check.py docstring updated to claude-opus-5

Verification:
- ✅ python3 -m py_compile: all 11 modules pass
- ✅ nix flake check: passes

.github/workflows/deploy.yml

@@ -26,7 +26,7 @@ jobs:
 
       - name: Python syntax check
         run: |
-          python3 -m py_compile run_server.py auto_merge_bot.py callback_server.py setup_app.py setup_all.py generate_manifest.py generate_manifest_domain.py start_server.py
+          python3 -m py_compile src/run_server.py src/auto_merge_bot.py src/callback_server.py scripts/setup_app.py scripts/setup_all.py scripts/generate_manifest.py scripts/generate_manifest_domain.py src/start_server.py src/sync-key.py src/health-check.py src/trivial_merge.py
 
   build-and-deploy:
     needs: syntax-check

.gitignore

@@ -5,13 +5,20 @@ webhook_secret.txt
 whsec*.txt
 .secrets.toml
 .env
+.env.*
 
 # Python
 __pycache__/
 *.pyc
 result
+
+# Packaging
 *.egg-info/
 .eggs/
 dist/
 build/
 .venv/
+
+# Runtime data
+*.log
+*.pid

CONTRIBUTING.md

@@ -0,0 +1,29 @@
+# Contributing to PR-Agent Server
+
+## Development Workflow
+
+1. Fork the repo
+2. Create a feature branch: `git checkout -b feat/your-feature`
+3. Make changes — keep files organized in the project layout:
+   - `src/` for application modules
+   - `scripts/` for setup/deployment helpers
+   - `templates/` for config templates
+4. Syntax check: `python3 -m py_compile src/*.py scripts/*.py`
+5. Nix build: `nix build .#default` (verify the flake still builds)
+6. Commit with descriptive message + push
+7. Open PR — the server's auto-merge bot will review it
+
+## Standards
+
+- **Python**: 4-space indent, type hints where practical, no hardcode secrets
+- **Secrets**: Always via environment variables or BWS at runtime — never in source
+- **Model names**: Must be tested live against 9router before committing (strip `openai/` prefix issue)
+- **Nix**: Update `flake.nix` `installPhase` if you move files between directories
+
+## Testing Checklist
+
+- [ ] `python3 -m py_compile` passes on all modified files
+- [ ] `nix build .#default` succeeds
+- [ ] CI syntax-check job passes
+- [ ] New models tested via curl to 9router (not assumed)
+- [ ] No secret values in git history (`sk-[a-z0-9]+` patterns)

README.md

@@ -1,55 +1,102 @@
 # PR-Agent Server
 
-Nix-deployed GitHub App server for automated PR review + auto-merge, using a custom LLM endpoint (9router/Omniroute).
+Nix-deployed GitHub App server for **automated PR review + auto-merge** using a custom LLM endpoint (9router/Omniroute).
 
 ## Architecture
 
 ```
-GitHub webhook → Cloudflare DNS → Caddy (reverse proxy, :4002)
+GitHub webhook → Caddy (reverse proxy, :4002)
-    → pr-agent-server (Nix profile, uvicorn)
+    → pr-agent-server (Nix profile, uvicorn on :4002)
         → PR-Agent github_app.py (FastAPI)
             → 9router API (custom OpenAI-compatible endpoint)
 ```
 
-## Components
+## Project Layout
 
-| File | Purpose |
+```
-|------|---------|
+pr-agent-server/
-| `run_server.py` | Main FastAPI server — GitHub App webhook handler + analytics/metrics + Discord notifications |
+├── src/                    # Main application modules
-| `auto_merge_bot.py` | Periodic bot that finds reviewed PRs and merges them |
+│   ├── run_server.py       # FastAPI server + analytics/metrics + Discord webhook
-| `trivial_merge.py` | Trivial PR fast-path (docs-only, dependabot, <100 lines) |
+│   ├── auto_merge_bot.py   # Periodic PR review→approve→merge bot
-| `health-check.py` | Model health watchdog — tests primary + fallback models against 9router |
+│   ├── trivial_merge.py    # Trivial PR fast-path (docs/dependabot/tiny diffs)
-| `sync-key.py` | Auto-syncs the BWS router key to disk on every service start |
+│   ├── health-check.py     # Model health watchdog (tests primary + fallbacks)
-| `generate_manifest.py` | Creates GitHub App manifest URL |
+│   ├── sync-key.py         # Auto-syncs BWS router key to disk on service start
-| `callback_server.py` | Dev callback server for receiving GitHub App credentials |
+│   ├── callback_server.py  # Dev callback server for GitHub App manifest
-| `flake.nix` | Nix build definition — produces the deployable package |
+│   ├── start_server.py     # Legacy server start script
+│   └── config/             # Runtime config (gitignored at deploy time)
+├── scripts/                # Setup and deployment helpers
+│   ├── setup_all.py        # Full setup: manifest + config + systemd service
+│   ├── setup_app.py        # App-specific setup
+│   ├── generate_manifest.py  # GitHub App manifest URL generator
+│   └── generate_manifest_domain.py
+├── templates/
+│   └── manifest.json       # GitHub App manifest template
+├── .github/workflows/
+│   ├── deploy.yml          # CI: syntax → build → deploy → GC
+│   ├── flakehub-publish-rolling.yaml
+│   └── mirror-gitea.yml
+├── flake.nix               # Nix build (creates venv + binary wrappers)
+├── flake.lock              # Pinned Nix dependencies
+├── .editorconfig           # Editor formatting rules
+├── .gitignore
+└── README.md
+```
 
 ## Development
 
-```bash
+### Prerequisites
-# Syntax check all Python files
+- Nix (for builds)
-python3 -m py_compile run_server.py auto_merge_bot.py health-check.py sync-key.py trivial_merge.py callback_server.py generate_manifest.py setup_all.py setup_app.py start_server.py
+- Python 3.12+
+- GitHub App credentials (App ID, private key, webhook secret)
+- BWS (Bitwarden Secrets Manager) access token
 
-# Build with Nix
+### Local testing
+
+```bash
+# Syntax check
+python3 -m py_compile src/run_server.py src/auto_merge_bot.py src/health-check.py src/sync-key.py src/trivial_merge.py src/callback_server.py
+
+# Nix build
 nix build .#default
 
-# Deploy (CI does this automatically on push to main)
+# Run server (after setting up secrets)
-nix copy --to ssh://user@vps $STORE_PATH
+export BWS_ACCESS_TOKEN="<your-bws-token>"
-ssh user@vps "sudo /nix/var/nix/profiles/default/bin/nix-env --profile /nix/var/nix/profiles/pr-agent-server --set '$STORE_PATH'"
+nix run .#pr-agent-server-sync-key  # syncs the router key
-ssh user@vps "sudo systemctl restart pr-agent-server"
+nix run .#pr-agent-server            # starts uvicorn on :3000
+
+# Health check
+nix run .#pr-agent-server-health-check
 ```
 
+## Deployment
+
+Deploy is fully automated via GitHub Actions on push to `main`:
+
+```yaml
+# .github/workflows/deploy.yml
+1. syntax-check  → python3 py_compile all modules
+2. build-and-deploy → nix build → SSH to VPS → update profile → restart service
+3. cleanup → nix-gc-vps.sh (with profile link repair)
+```
+
+Secrets required in GitHub Actions:
+- `VPS_HOST` — VPS IP address
+- `VPS_USER` — SSH user
+- `SSH_PRIVATE_KEY` — SSH private key for deploy user
+- `GITEA_TOKEN` — for Gitea mirror (if using mirror workflow)
+
 ## Ops
 
-- **Health watchdog**: cron `pr-agent-health-watchdog` (every 10m) → `~/.hermes/scripts/pr-agent-health-check.sh`
+- **Health watchdog**: cron `pr-agent-health-watchdog` (every 10 min) → `~/.hermes/scripts/pr-agent-health-check.sh` → Nix binary `pr-agent-health-check`
-- **Key sync**: systemd `ExecStartPre` → `pr-agent-sync-key` (BWS → disk)
+- **Key auto-sync**: systemd `ExecStartPre=/usr/local/bin/bws-exec pr-agent -- <profile>/bin/pr-agent-sync-key`
-- **Prometheus**: `GET /api/metrics` → `pr_agent_requests_total`, `pr_agent_requests_by_command`, `pr_agent_model_failures`
+- **Prometheus**: `GET /api/metrics` → `pr_agent_requests_total`, `pr_agent_model_failures`
-- **Analytics**: `GET /api/analytics` → JSON summary of recent events + failures
+- **Analytics**: `GET /api/analytics` → JSON summary (unwrap `"record"` field)
-- **Discord**: `POST /api/v1/notify_review` → webhook delivery for review complete/failed
+- **Discord**: `POST /api/v1/notify_review` → pr-agent-ops webhook
 
 ## Nix Profile Integrity
 
-After deploy, run `nix-gc-vps.sh` to clean up old generations. The GC script now includes
+⚠️ See the `devops/pr-agent-deployment` skill for troubleshooting broken `-link` profile symlinks after `nix store gc`. The GC script (`/usr/local/bin/nix-gc-vps.sh`) now includes a repair step.
-a repair step that fixes broken profile symlinks before running `nix store gc`, preventing
+
-the issue where profile `-link` dirs get deleted and store paths become unreferenced (see
+## License
-`devops/pr-agent-deployment` skill for full troubleshooting).
+
+MIT — see [LICENSE](LICENSE) if present at deploy.

flake.nix

@@ -38,34 +38,50 @@
 
           installPhase = ''
             mkdir -p $out/bin $out/lib/pr-agent-server
-            cp run_server.py $out/lib/pr-agent-server/
-            cp sync-key.py $out/lib/pr-agent-server/
-            cp health-check.py $out/lib/pr-agent-server/
-            cp trivial_merge.py $out/lib/pr-agent-server/
-            cp auto_merge_bot.py $out/lib/pr-agent-server/
 
+            # Copy server modules
+            cp src/run_server.py $out/lib/pr-agent-server/
+            cp src/sync-key.py $out/lib/pr-agent-server/
+            cp src/health-check.py $out/lib/pr-agent-server/
+            cp src/trivial_merge.py $out/lib/pr-agent-server/
+            cp src/auto_merge_bot.py $out/lib/pr-agent-server/
+            cp src/callback_server.py $out/lib/pr-agent-server/
+
+            # Wrapper: pr-agent-server (main FastAPI webhook server)
             cat > $out/bin/pr-agent-server << WRAPPER
 #!${pkgs.runtimeShell}
-export PATH=${pkgs.git}/bin:\$PATH
+export PATH=${pkgs.git}/bin:$PATH
-export LD_LIBRARY_PATH=${pkgs.stdenv.cc.cc.lib}/lib:\$LD_LIBRARY_PATH
+export LD_LIBRARY_PATH=${pkgs.stdenv.cc.cc.lib}/lib:$LD_LIBRARY_PATH
 cd $out/lib/pr-agent-server
 exec $out/venv/bin/python run_server.py
 WRAPPER
             chmod +x $out/bin/pr-agent-server
 
+            # Wrapper: pr-agent-sync-key (BWS key sync)
             cat > $out/bin/pr-agent-sync-key << WRAPPER2
 #!${pkgs.runtimeShell}
-export LD_LIBRARY_PATH=${pkgs.stdenv.cc.cc.lib}/lib:\$LD_LIBRARY_PATH
+export LD_LIBRARY_PATH=${pkgs.stdenv.cc.cc.lib}/lib:$LD_LIBRARY_PATH
 exec $out/venv/bin/python $out/lib/pr-agent-server/sync-key.py
 WRAPPER2
             chmod +x $out/bin/pr-agent-sync-key
 
+            # Wrapper: pr-agent-health-check (model health watchdog)
             cat > $out/bin/pr-agent-health-check << WRAPPER3
 #!${pkgs.runtimeShell}
-export LD_LIBRARY_PATH=${pkgs.stdenv.cc.cc.lib}/lib:\$LD_LIBRARY_PATH
+export LD_LIBRARY_PATH=${pkgs.stdenv.cc.cc.lib}/lib:$LD_LIBRARY_PATH
 exec $out/venv/bin/python $out/lib/pr-agent-server/health-check.py
 WRAPPER3
             chmod +x $out/bin/pr-agent-health-check
+
+            # Wrapper: pr-agent-auto-merge (merge worker)
+            cat > $out/bin/pr-agent-auto-merge << WRAPPER4
+#!${pkgs.runtimeShell}
+export PATH=${pkgs.git}/bin:$PATH
+export LD_LIBRARY_PATH=${pkgs.stdenv.cc.cc.lib}/lib:$LD_LIBRARY_PATH
+cd $out/lib/pr-agent-server
+exec $out/venv/bin/python auto_merge_bot.py
+WRAPPER4
+            chmod +x $out/bin/pr-agent-auto-merge
           '';
         };
       });

manifest_current.json

@@ -1,24 +0,0 @@
-{
-  "name": "pr-agent-auto-review",
-  "url": "https://github.com/asepharyana",
-  "hook_attributes": {
-    "url": "https://pr-agent.asepharyana.my.id/api/v1/github_webhooks",
-    "active": true
-  },
-  "redirect_url": "https://pr-agent.asepharyana.my.id/setup/callback",
-  "callback_urls": [
-    "https://pr-agent.asepharyana.my.id/setup/callback"
-  ],
-  "public": false,
-  "default_events": [
-    "pull_request",
-    "issue_comment"
-  ],
-  "default_permissions": {
-    "pull_requests": "write",
-    "issues": "write",
-    "contents": "read",
-    "metadata": "read",
-    "checks": "write"
-  }
-}

manifest_final.json

@@ -1,24 +0,0 @@
-{
-  "name": "pr-agent-auto-review",
-  "url": "https://github.com/asepharyana",
-  "hook_attributes": {
-    "url": "https://pr-agent.asepharyana.my.id/api/v1/github_webhooks",
-    "active": true
-  },
-  "redirect_url": "https://pr-agent.asepharyana.my.id/setup/callback",
-  "callback_urls": [
-    "https://pr-agent.asepharyana.my.id/setup/callback"
-  ],
-  "public": false,
-  "default_events": [
-    "pull_request",
-    "issue_comment"
-  ],
-  "default_permissions": {
-    "pull_requests": "write",
-    "issues": "write",
-    "contents": "read",
-    "metadata": "read",
-    "checks": "write"
-  }
-}

src/health-check.py

@@ -78,7 +78,7 @@ def check_model(model: str, key: str) -> tuple:
     """Returns (ok: bool, detail: str). Uses raw HTTP (no litellm dependency).
 
     NOTE: litellm strips the 'openai/' provider prefix before sending the
-    request body. 9router resolves bare aliases (e.g. 'claude-opus-4-8') to
+    request body. 9router resolves bare aliases (e.g. 'claude-opus-5') to
     its own routing; WITH the prefix it tries the 'openai' provider upstream,
     which has no credentials → 404 'No active credentials for provider: openai'.
     So we strip the prefix here to mirror exactly what the server sends.