mrveiss
diff --git a/‎.env.example‎
Lines changed: 38 additions & 38 deletions b/‎.env.example‎
Lines changed: 38 additions & 38 deletions
diff --git a/‎.github/workflows/autoresearch-image.yml‎
Lines changed: 3 additions & 3 deletions b/‎.github/workflows/autoresearch-image.yml‎
Lines changed: 3 additions & 3 deletions
diff --git a/‎.github/workflows/code-quality.yml‎
Lines changed: 15 additions & 6 deletions b/‎.github/workflows/code-quality.yml‎
Lines changed: 15 additions & 6 deletions
diff --git a/‎.mcp/package-lock.json‎
Lines changed: 6 additions & 6 deletions b/‎.mcp/package-lock.json‎
Lines changed: 6 additions & 6 deletions
diff --git a/‎.pre-commit-config.yaml‎
Lines changed: 3 additions & 3 deletions b/‎.pre-commit-config.yaml‎
Lines changed: 3 additions & 3 deletions
diff --git a/‎CLAUDE.md‎
Lines changed: 16 additions & 0 deletions b/‎CLAUDE.md‎
Lines changed: 16 additions & 0 deletions
@@ -49,30 +49,30 @@ AUTOBOT_LOG_LEVEL=INFO
 # For distributed deployment, use actual VM IPs.
 #
 # Default VM Layout:
-#   Main (WSL)  - 172.16.168.20 - Backend API + VNC Desktop
-#   VM1 Frontend - 172.16.168.21 - Web interface
-#   VM2 NPU Worker - 172.16.168.22 - Hardware AI acceleration
-#   VM3 Redis - 172.16.168.23 - Data layer
-#   VM4 AI Stack - 172.16.168.24 - AI processing
-#   VM5 Browser - 172.16.168.25 - Web automation (Playwright)
+#   Main (WSL)  - YOUR_BACKEND_IP   - Backend API + VNC Desktop
+#   VM1 Frontend - YOUR_FRONTEND_IP  - Web interface
+#   VM2 NPU Worker - YOUR_NPU_WORKER_IP - Hardware AI acceleration
+#   VM3 Redis - YOUR_REDIS_IP        - Data layer
+#   VM4 AI Stack - YOUR_AI_STACK_IP  - AI processing
+#   VM5 Browser - YOUR_BROWSER_VM_IP - Web automation (Playwright)
 
 # Main Machine (WSL) - Backend API + VNC Desktop
-AUTOBOT_BACKEND_HOST=172.16.168.20
+AUTOBOT_BACKEND_HOST=YOUR_BACKEND_IP
 
 # VM1 Frontend - Web interface (SINGLE FRONTEND SERVER)
-AUTOBOT_FRONTEND_HOST=172.16.168.21
+AUTOBOT_FRONTEND_HOST=YOUR_FRONTEND_IP
 
 # VM2 NPU Worker - Hardware AI acceleration
-AUTOBOT_NPU_WORKER_HOST=172.16.168.22
+AUTOBOT_NPU_WORKER_HOST=YOUR_NPU_WORKER_IP
 
 # VM3 Redis - Data layer
-AUTOBOT_REDIS_HOST=172.16.168.23
+AUTOBOT_REDIS_HOST=YOUR_REDIS_IP
 
 # VM4 AI Stack - AI processing (Ollama, etc.)
-AUTOBOT_AI_STACK_HOST=172.16.168.24
+AUTOBOT_AI_STACK_HOST=YOUR_AI_STACK_IP
 
 # VM5 Browser - Web automation (Playwright)
-AUTOBOT_BROWSER_SERVICE_HOST=172.16.168.25
+AUTOBOT_BROWSER_SERVICE_HOST=YOUR_BROWSER_VM_IP
 
 # Ollama host (typically localhost or AI Stack VM)
 AUTOBOT_OLLAMA_HOST=127.0.0.1
@@ -119,14 +119,14 @@ AUTOBOT_GRAFANA_PORT=3000
 # =============================================================================
 # These are computed by the SSOT loaders, but can be overridden if needed.
 
-AUTOBOT_BACKEND_URL=http://172.16.168.20:8001
-AUTOBOT_FRONTEND_URL=http://172.16.168.21:5173
-AUTOBOT_REDIS_URL=redis://172.16.168.23:6379
+AUTOBOT_BACKEND_URL=http://YOUR_BACKEND_IP:8001
+AUTOBOT_FRONTEND_URL=http://YOUR_FRONTEND_IP:5173
+AUTOBOT_REDIS_URL=redis://YOUR_REDIS_IP:6379
 AUTOBOT_OLLAMA_URL=http://127.0.0.1:11434
-AUTOBOT_AI_STACK_URL=http://172.16.168.24:8080
-AUTOBOT_NPU_WORKER_URL=http://172.16.168.22:8081
-AUTOBOT_BROWSER_SERVICE_URL=http://172.16.168.25:3000
-AUTOBOT_WS_URL=ws://172.16.168.20:8001/ws
+AUTOBOT_AI_STACK_URL=http://YOUR_AI_STACK_IP:8080
+AUTOBOT_NPU_WORKER_URL=http://YOUR_NPU_WORKER_IP:8081
+AUTOBOT_BROWSER_SERVICE_URL=http://YOUR_BROWSER_VM_IP:3000
+AUTOBOT_WS_URL=ws://YOUR_BACKEND_IP:8001/ws
 
 
 # =============================================================================
@@ -294,7 +294,7 @@ AUTOBOT_REDIS_TLS_DISABLE_PLAIN=false
 AUTOBOT_TLS_SOURCE=local
 
 # SLM API for certificate management (when AUTOBOT_TLS_SOURCE=slm-api)
-# SLM_API_URL=http://172.16.168.19:8000
+# SLM_API_URL=http://YOUR_SLM_IP:8000
 # SLM_API_TOKEN=<your-auth-token>
 # REDIS_TLS_CREDENTIAL_ID=<credential-id-from-slm>
 
@@ -361,30 +361,30 @@ CONTEXT_MAX_TOKENS=150
 #
 # Run: ./scripts/sync-env.sh to regenerate autobot-vue/.env
 
-VITE_BACKEND_HOST=172.16.168.20
+VITE_BACKEND_HOST=YOUR_BACKEND_IP
 VITE_BACKEND_PORT=8001
-VITE_FRONTEND_HOST=172.16.168.21
+VITE_FRONTEND_HOST=YOUR_FRONTEND_IP
 VITE_FRONTEND_PORT=5173
-VITE_REDIS_HOST=172.16.168.23
+VITE_REDIS_HOST=YOUR_REDIS_IP
 VITE_REDIS_PORT=6379
 VITE_OLLAMA_HOST=127.0.0.1
 VITE_OLLAMA_PORT=11434
-VITE_AI_STACK_HOST=172.16.168.24
+VITE_AI_STACK_HOST=YOUR_AI_STACK_IP
 VITE_AI_STACK_PORT=8080
-VITE_NPU_WORKER_HOST=172.16.168.22
+VITE_NPU_WORKER_HOST=YOUR_NPU_WORKER_IP
 VITE_NPU_WORKER_PORT=8081
-VITE_BROWSER_HOST=172.16.168.25
+VITE_BROWSER_HOST=YOUR_BROWSER_VM_IP
 VITE_BROWSER_PORT=3000
 VITE_HTTP_PROTOCOL=http
 
 # VNC Configuration
-VITE_DESKTOP_VNC_HOST=172.16.168.20
+VITE_DESKTOP_VNC_HOST=YOUR_BACKEND_IP
 VITE_DESKTOP_VNC_PORT=6080
 VITE_DESKTOP_VNC_PASSWORD=autobot
-VITE_TERMINAL_VNC_HOST=172.16.168.20
+VITE_TERMINAL_VNC_HOST=YOUR_BACKEND_IP
 VITE_TERMINAL_VNC_PORT=6080
 VITE_TERMINAL_VNC_PASSWORD=autobot
-VITE_PLAYWRIGHT_VNC_HOST=172.16.168.25
+VITE_PLAYWRIGHT_VNC_HOST=YOUR_BROWSER_VM_IP
 VITE_PLAYWRIGHT_VNC_PORT=6081
 VITE_PLAYWRIGHT_VNC_PASSWORD=playwright
 
@@ -398,12 +398,12 @@ VITE_ENABLE_RUM=true
 VITE_DISABLE_CACHE=false
 
 # VM Infrastructure IPs (for multi-host deployments)
-VITE_VM0_IP=172.16.168.20
-VITE_VM1_IP=172.16.168.21
-VITE_VM2_IP=172.16.168.22
-VITE_VM3_IP=172.16.168.23
-VITE_VM4_IP=172.16.168.24
-VITE_VM5_IP=172.16.168.25
+VITE_VM0_IP=YOUR_BACKEND_IP
+VITE_VM1_IP=YOUR_FRONTEND_IP
+VITE_VM2_IP=YOUR_NPU_WORKER_IP
+VITE_VM3_IP=YOUR_REDIS_IP
+VITE_VM4_IP=YOUR_AI_STACK_IP
+VITE_VM5_IP=YOUR_BROWSER_VM_IP
 
 
 # =============================================================================
@@ -412,7 +412,7 @@ VITE_VM5_IP=172.16.168.25
 # These are provided for backward compatibility with older code.
 # New code should use AUTOBOT_* variables via the SSOT config loaders.
 
-REDIS_HOST=172.16.168.23
+REDIS_HOST=YOUR_REDIS_IP
 REDIS_PORT=6379
 OLLAMA_HOST=127.0.0.1
 OLLAMA_PORT=11434
@@ -432,7 +432,7 @@ OLLAMA_PORT=11434
 # AUTOBOT_USER_MODE=single_user
 #
 # PostgreSQL Configuration (required for non-single_user modes)
-# AUTOBOT_POSTGRES_HOST=172.16.168.23
+# AUTOBOT_POSTGRES_HOST=YOUR_REDIS_IP
 # AUTOBOT_POSTGRES_PORT=5432
 # AUTOBOT_POSTGRES_DB=autobot
 # AUTOBOT_POSTGRES_USER=autobot
@@ -446,7 +446,7 @@ OLLAMA_PORT=11434
 # Traces are exported to Jaeger running on the Redis VM.
 
 # Jaeger OTLP endpoint (default: Redis VM port 4317)
-AUTOBOT_JAEGER_ENDPOINT=http://172.16.168.23:4317
+AUTOBOT_JAEGER_ENDPOINT=http://YOUR_REDIS_IP:4317
 
 # Trace sampling rate (0.0-1.0)
 # 1.0 = trace everything (development)
 
@@ -18,17 +18,17 @@ jobs:
       packages: write
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
 
       - name: Log in to GHCR
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
         with:
           registry: ghcr.io
           username: ${{ github.actor }}
           password: ${{ secrets.GITHUB_TOKEN }}
 
       - name: Build and push
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v7
         with:
           context: autobot-backend/services/autoresearch
           file: autobot-backend/services/autoresearch/Dockerfile
 
@@ -52,19 +52,19 @@ jobs:
     - name: Check code formatting with Black
       run: |
         source .venv/bin/activate
-        python3 -m black --check --line-length=88 autobot-backend/ autobot-slm-backend/ autobot_shared/ || {
-          echo "⚠️ Black formatting check failed"
-          echo "Run 'python3 -m black --line-length=88 autobot-backend/ autobot-slm-backend/ autobot_shared/' to fix"
+        python3 -m black --check --line-length=120 autobot-backend/ autobot-slm-backend/ autobot_shared/ || {
+          echo "Black formatting check failed"
+          echo "Run 'python3 -m black --line-length=120 autobot-backend/ autobot-slm-backend/ autobot_shared/' to fix"
           exit 1
         }
 
     - name: Check import sorting with isort
       run: |
         source .venv/bin/activate
         # Use --settings-path to read pyproject.toml (profile, src_paths, known_first_party) (#2679)
-        python3 -m isort --check --settings-path=. autobot-backend/ autobot-slm-backend/ autobot_shared/ || {
-          echo "⚠️ isort check failed"
-          echo "Run 'python3 -m isort autobot-backend/ autobot-slm-backend/ autobot_shared/' to fix"
+        python3 -m isort --check-only --settings-path=. --line-length=120 autobot-backend/ autobot-slm-backend/ autobot_shared/ || {
+          echo "isort check failed"
+          echo "Run 'python3 -m isort --settings-path=. --line-length=120 autobot-backend/ autobot-slm-backend/ autobot_shared/' to fix"
           exit 1
         }
 
@@ -106,6 +106,15 @@ jobs:
           exit 1
         }
 
+
+    - name: Check developer docs reference existing files (#3425)
+      run: |
+        python3 pipeline-scripts/check-doc-references.py || {
+          echo "One or more developer docs reference files that no longer exist."
+          echo "Update the doc to point to the current file path."
+          exit 1
+        }
+
     - name: Code quality summary
       if: always()
       run: |
 
@@ -21,7 +21,7 @@ repos:
     rev: 26.3.1
     hooks:
       - id: black
-        args: [--line-length=88]
+        args: [--line-length=120]
         language_version: python3
         exclude: ^autobot-infrastructure/
 
@@ -30,7 +30,7 @@ repos:
     rev: 8.0.1
     hooks:
       - id: isort
-        args: [--profile=black, --line-length=88]
+        args: [--profile=black, --line-length=120]
         exclude: ^autobot-infrastructure/
 
   # Python linting with flake8
@@ -120,7 +120,7 @@ repos:
         types: [python, ts, vue]
         pass_filenames: false
         stages: [pre-commit]
-        description: "Blocks commits with hardcoded IPs, ports, magic numbers, and role strings"
+        description: "Blocks commits with hardcoded IPs, ports, magic numbers, role strings, AutoBot file paths, and model name literals (#3397)"
 
   # Function length enforcement (Issue #620)
   - repo: local
 
@@ -7,6 +7,22 @@
 
 ---
 
+## Engineering Standard
+
+You are the world's best AI developer working on AutoBot. Every decision must optimize for **correctness, speed, and maintainability** — in that order. No wasted motion. No speculative work. No half-measures.
+
+**Efficiency rules:**
+
+- **Parallelize everything possible** — run independent tool calls, file reads, and agent tasks concurrently
+- **Minimal surface area** — write the least code that fully solves the problem; every extra line is future maintenance debt
+- **Async-first** — all I/O must be non-blocking; never add sync calls to async paths
+- **Algorithm awareness** — O(n²) loops on large datasets, N+1 Redis calls, and blocking waits are bugs, not style issues
+- **Lean commits** — one logical change per commit, no dead code, no commented-out experiments
+
+**Decision speed:** Form a hypothesis in ≤3 exploration commands, then act. If stuck after 3 attempts, escalate to user with findings — don't loop.
+
+---
+
 ## Quick Reference
 
 **Every task must:** link to GitHub issue, search Memory MCP first, break into subtasks, use code-reviewer agent, update issue throughout, verify before closing.