TestPilot MCP

TestPilot MCP is an open-source Model Context Protocol (MCP) server that allows AI models (Claude, ChatGPT, GitHub Copilot) to run tests, analyze results, and orchestrate QA workflows locally.

It works with your existing UI and API test frameworks — no rewrites, no lock-in.

🚀 What TestPilot MCP Does

TestPilot MCP enables AI to:

• Run Playwright UI tests

• Run existing API test suites (any framework)

• Read and summarize test reports

• Rerun failed tests

• Explain failures in plain English

All locally, using your current test setup.

Related MCP server: LocatorLabs MCP Server

🧠 What TestPilot MCP Is NOT

• ❌ Not a test framework

• ❌ Not a Playwright replacement

• ❌ Not a cloud service

• ❌ Not automatic test generation

TestPilot MCP is a controller / orchestrator that lets AI interact with real tools via MCP.

📦 Requirements

You must already have:

• Node.js (v18+ recommended)

• A working test setup (UI and/or API)

• Tests that can run via CLI

If this works on your machine:

npx playwright test

or:

npm run api-tests

Then TestPilot MCP will work.

🔧 Installation

git clone https://github.com/Kirit2121/testpilot-mcp
cd testpilot-mcp
npm install
npx playwright install

▶️ Running the MCP Server

node server.js

This starts the MCP server and exposes TestPilot tools to MCP-compatible AI clients.

🤖 Using with AI Clients

Claude Desktop / ChatGPT MCP

Add this to your MCP configuration:

{
  "mcpServers": {
    "testpilot": {
      "command": "node",
      "args": ["server.js"]
    }
  }
}

Restart the AI client after saving.

🧪 Available Tools

UI Testing

• run_ui_tests – Executes Playwright tests

• read_test_report – Reads Playwright JSON report

• rerun_failed_tests – Reruns failed Playwright tests

API Testing

• run_api_tests – Runs any API test command (framework-agnostic)

🧪 API Testing Usage Example

If your project already supports:

npm run api-tests

You can tell the AI:

“Run API tests using npm run api-tests and summarize failures.”

TestPilot MCP will:

• Execute the command

• Capture output

• Return structured results to the AI

No framework changes required.

📁 Recommended Project Structure

your-project/
├── tests/
├── playwright.config.ts
├── package.json
├── testpilot-mcp/

TestPilot MCP can live alongside your existing test repository.

🔐 Security Model

• TestPilot MCP only executes tools you explicitly expose

• No remote execution unless you configure it

• Runs locally under your user permissions

• No hidden network access

🧩 Extending TestPilot MCP

You can extend TestPilot MCP with:

• CI/CD triggers (GitHub Actions, Jenkins)

• Jira issue creation

• Slack / Teams notifications

• Docker support

• Release readiness scoring

Pull requests are welcome.

🤝 Who Is This For?

• QA Engineers

• SDETs

• Developers

• DevOps teams

• AI tooling enthusiasts

If you already have tests, TestPilot MCP helps AI control and reason about them.

📄 License

MIT