Healenium MCP

Use your Healenium data inside AI assistants (Cursor, Claude Desktop, GitHub Copilot, and other MCP‑compatible clients).

What is Healenium MCP?

The Healenium MCP Server is a bridge between your Healenium instance and AI chat tools that support the Model Context Protocol (MCP). Once the server is running and connected to your AI client, you can ask questions in plain language about your test runs, reports, and selectors, and get answers based on live data from Healenium.

Prerequisites

• A running Healenium instance (backend URL reachable from your machine).
• An MCP‑compatible AI client (e.g. Cursor, Claude Desktop).
• Either Docker (to run the official image) or a built binary of the Healenium MCP server.

You do not need to start the server manually: the client starts it when needed using the configuration below.

Client configuration

Add the Healenium MCP server in your client’s MCP settings. The client will run the server process when you use the AI.

Option 1: Docker (recommended)

Use the official image and pass your Healenium URL via environment. Example for Cursor (~/.cursor/mcp.json or Cursor Settings → MCP):

{
  "mcpServers": {
    "healenium": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e", "HL_URL",
        "healenium/healenium-mcp-server:latest"
      ],
      "env": {
        "HL_URL": "http://localhost:7878"
      }
    }
  }
}

Replace http://localhost:7878 with your Healenium backend URL (e.g. https://healenium.example.com or http://healenium:7878 if Healenium runs in another container).

Optional: to keep a persistent UID across restarts, add a volume in args: -v, healenium-mcp-data:/server.

Option 2: Binary (no Docker)

If you have the healenium-mcp-server binary (from releases or built from source), point the client to it and pass the Healenium URL.

Cursor — in MCP settings add a server entry like:

{
  "mcpServers": {
    "healenium": {
      "command": "/path/to/healenium-mcp-server",
      "args": ["--service-url", "http://localhost:7878"],
      "env": {}
    }
  }
}

Or pass the URL via environment (the binary respects HL_URL):

"healenium": {
  "command": "/path/to/healenium-mcp-server",
  "env": {
    "HL_URL": "http://localhost:7878"
  }
}

Replace /path/to/healenium-mcp-server with the actual path and the URL with your Healenium backend.

Claude Desktop — in the MCP config file use the same structure: command (path to binary or docker), args if needed, and env with HL_URL.

Server configuration reference

These options apply when the server is run (by the client or by you). Environment variables and CLI flags:

Available tools

Once the MCP server is connected to your AI client, the following tools are available so the assistant can query and update Healenium on your behalf.

Other clients

• Claude Desktop: Edit the MCP config file (e.g. claude_desktop_config.json) and add a server entry with the same command / args / env structure as above.
• Other MCP clients: Use your client’s “MCP server” or “Custom server” settings; provide the same command (Docker or binary) and HL_URL. For the protocol, see the official MCP documentation.