httprunner

HttpRunner MCP server exposes UI automation capabilities as part of the HttpRunner ecosystem. This MCP server is designed to integrate with the HttpRunner tooling stack to manage, orchestrate, and expose UI automation workflows across devices and platforms. After starting the MCP server, you can use the accompanying MCP tooling (mcphost, mcp-server) to interact with UI automation tasks, launch automated UI test scenarios, and retrieve status or results from running experiments. The server is intended to bridge the MCP protocol with HttpRunner's UI automation features, enabling you to define test scenarios, trigger executions, and collect reports in a centralized manner.

Prerequisites:

• Docker installed and running, or a Go development environment if you prefer building locally.
• Network access to pull required images or modules.

Option A: Run via Docker (recommended)

1. Pull and run the HttpRunner MCP server image docker pull httprunner/httprunner:latest docker run -d --name hrp-mcp -p 8080:8080 httprunner/httprunner:latest
2. Verify the server is up curl http://localhost:8080/health

Option B: Build from source (Go)

1. Install Go (1.20+)
2. Clone the repository and navigate to the project root
3. Build the server go build ./...
4. Run the server ./httprunner-mcp-server

Option C: Alternative runtimes

• If you prefer using a different runtime, ensure the server exposes the MCP API endpoints over HTTP/HTTPS and that you configure your MCP client to target the running server address and port.

Configuration:

• Ensure the server is reachable at the address used by your MCP clients (default port is 8080 unless overridden).
• If authentication is required, configure the appropriate environment variables or command-line flags as documented by HttpRunner MCP.

Tips and common issues:

• Ensure network access between the MCP client and the HttpRunner MCP server; firewalls may block the required port.
• If you see health check failures, verify that the container has started correctly or that the Go binary has been built with compatible dependencies.
• When running in Docker, consider mounting a volume for persistent logs and reports if your workflow requires audit trails.
• Review HttpRunner documentation for UI automation capabilities (Android/iOS/Web) to understand the available test scenarios and reporting formats.
• If you upgrade HttpRunner, verify that the MCP API endpoints remain compatible; some fields or endpoints may evolve between versions.
• Environment variables to consider (placeholders):
  • HRP_MCP_API_KEY: API key for MCP authentication
  • HRP_MCP_PORT: Port on which the MCP server listens
  • HRP_LOG_LEVEL: Logging level (e.g., INFO, DEBUG)