Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@BarrHawk test the login flow on example.com"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
██████╗ █████╗ ██████╗ ██████╗ ██╗ ██╗ █████╗ ██╗ ██╗██╗ ██╗
██╔══██╗██╔══██╗██╔══██╗██╔══██╗██║ ██║██╔══██╗██║ ██║██║ ██╔╝
██████╔╝███████║██████╔╝██████╔╝███████║███████║██║ █╗ ██║█████╔╝
██╔══██╗██╔══██║██╔══██╗██╔══██╗██╔══██║██╔══██║██║███╗██║██╔═██╗
██████╔╝██║ ██║██║ ██║██║ ██║██║ ██║██║ ██║╚███╔███╔╝██║ ██╗
╚═════╝ ╚═╝ ╚═╝╚═╝ ╚═╝╚═╝ ╚═╝╚═╝ ╚═╝╚═╝ ╚═╝ ╚══╝╚══╝ ╚═╝ ╚═╝
v0.5.0 | TRIPARTITE RELEASE"If your Agent can't test itself, it's just a hallucination."
Unified Stack - Single command bun run barrhawk starts everything
Integrated Dashboard - Live monitoring at http://localhost:7000/dashboard
Experience System - Self-healing selectors that learn across sessions
Tool Injection - New tools broadcast to running Igors in real-time
Lightning Strike - Igor auto-escalates to Claude when stuck
Hub Mode - Multi-Igor test orchestration with --hub flag
Related MCP server: chrometools-mcp
# 1. Clone and install
git clone git@github.com:barrhawk/barrhawk_premium_e2e_mcp.git
cd barrhawk_premium_e2e_mcp
bun install
# 2. Start the stack
bun run barrhawk
# 3. Submit a test
curl -X POST http://localhost:7001/plan \
-H "Content-Type: application/json" \
-d '{"intent":"take a screenshot of the homepage","url":"https://example.com"}'bun run barrhawk # Full tripartite stack (Bridge, Doctor, Igor, Frank)
bun run barrhawk --minimal # Just Bridge + Igor (lightweight)
bun run barrhawk --hub # Full stack + Test Orchestration HubFour specialized servers communicating via WebSocket:
Component | Port | Role |
Bridge | 7000 | Message bus, circuit breakers, dashboard |
Doctor | 7001 | Planner, intent parser, experience engine |
Igor | 7002 | Executor, Lightning Strike dual-mode |
Frankenstein | 7003 | Browser control, dynamic tool forge |
You send an intent to Doctor: "test the login flow"
Doctor consults experience, generates a plan, selects tools
Igor executes the plan step-by-step
Frankenstein controls the browser, takes screenshots
Experience records what worked for next time
Igor starts in "dumb mode" - fast, regex-based execution. When a step fails 3 times, Igor "strikes" - elevating to full Claude reasoning to solve the problem, then powers back down.
# Submit a plan
POST /plan
{"intent": "click the login button", "url": "https://example.com"}
# Check plan status
GET /plans
GET /plans/:id# Health check
GET /health
# Live dashboard
GET /dashboard# Status
GET /status
GET /health
# Current tool bag
GET /toolbag# Health and tools
GET /health
GET /tools/
├── bin/barrhawk.ts # CLI entry point
├── tripartite/ # The Core Stack
│ ├── bridge/ # Message Bus + Dashboard (7000)
│ ├── doctor/ # Planner + Swarm Logic (7001)
│ ├── igor/ # Executor + Lightning (7002)
│ ├── frankenstein/ # Browser + Tool Forge (7003)
│ └── shared/ # Common utilities
│ ├── experience.ts # Learning system
│ ├── tool-registry.ts # 120+ tool definitions
│ └── types.ts # Shared types
├── hub/ # Test Orchestration (--hub mode)
│ ├── index.ts # Hub API (7010)
│ ├── coordinator.ts # Multi-Igor sync (7011)
│ └── igor-db.ts # Database watcher (7012)
├── packages/ # Feature modules
│ ├── golden/ # Visual regression
│ ├── self-heal/ # Selector strategies
│ └── ... # 30+ packages
└── docs/ # Documentation
└── specs/ # Component specificationsBarrHawk exposes MCP tools for Claude Code, Cursor, and other AI assistants.
./scripts/generate-mcp-configs.sh// Natural language automation
frank_execute({ task: "Log into github.com", url: "https://github.com" })
// Parallel testing
frank_swarm_execute({ intent: "Test checkout flow", maxIgors: 4 })
// OS-level control (extensions, dialogs)
frank_os_mouse({ action: "click", x: 100, y: 200 })
frank_os_keyboard({ combo: "ctrl+shift+i" })
// Screenshots
frank_screenshot({ fullPage: true })BarrHawk learns from every test run:
Selectors: Which CSS selectors work for which elements
Timings: How long actions typically take per site
Errors: Common error patterns and their fixes
Sites: Known site patterns with pre-mapped selectors
Data stored in experiencegained/experience.json (configurable via EXPERIENCE_DIR).
# Run tests
bun test
# Type check
bun run typecheck
# View logs
tail -f /tmp/tripartite-*.logMIT
"Trust, but Verify." — BarrHawk
This server cannot be deployed
Browser-based QA for AI-built software. Test pages with real browsers via agents.
AI QA tester — real browsers scan sites for bugs, SEO, perf, and accessibility issues via chat.
Run multi-step tasks in a real Chrome browser: persistent environments, live view, human takeover.
AI QA that runs your app in a browser on every pull request: projects, test targets, test cases.
