home / mcp / tekmetric mcp server

Tekmetric MCP Server

MCP server for Tekmetric API

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "patclarke-mcp-tekmetric": {
      "url": "http://localhost:8080/sse",
      "headers": {
        "TEKMETRIC_API_KEY": "YOUR_TEKMETRIC_API_KEY"
      }
    }
  }
}

You deploy this MCP server to let an AI assistant read data from Tekmetric and perform automated, AI-assisted workflows. It enables you to retrieve appointment details, shop contact information, and repair order statuses in a read-only fashion, making your Tekmetric data accessible to your AI tools with secure, scoped access.

How to use

Connect your MCP client to the Tekmetric MCP server using the provided HTTP endpoint. The server exposes a read-only API that lets your AI assistant query appointment details, fetch shop information, and check the status of repair orders. You can design natural language prompts that map to these read-only actions, enabling conversational or voice-driven workflows without modifying Tekmetric data.

How to install

Prerequisites: you need Python installed (for local run) and optionally Docker for containerized execution. You also need a Tekmetric API key to authenticate requests.

# Prerequisites
# 1) Ensure Python is installed
python3 --version

# 2) Optionally install Docker for containerized runs
docker --version

# 3) Obtain Tekmetric API key from your Tekmetric account settings

Additional setup and run options

You can run the server locally with Python or run it via Docker. The server requires your Tekmetric API key to be available as an environment variable.

# Option A: Run locally with Python
# 1) Install dependencies
pip install -e .

# 2) Start the server on all interfaces port 8080
uvicorn mcp_tekmetric.servers.main:asgi_app --host 0.0.0.0 --port 8080

# Option B: Run with Docker
# Build the image
# docker build -t mcp-tekmetric .

# Run the container with your Tekmetric API key
# docker run -e TEKMETRIC_API_KEY=your_tekmetric_api_key -p 8080:8080 mcp-tekmetric

Health check and status

Verify that the server is running by querying the health endpoint.

curl http://localhost:8080/healthz
```
You should see:
```json
{"status": "ok"}

Connect your AI assistant

Configure your MCP client with the server’s HTTP endpoint to enable read-only interactions. The server URL below is the local development URL used in examples.

{
  "servers": {
    "local-tekmetric": {
      "type": "http",
      "url": "http://localhost:8080/sse",
      "args": []
    }
  }
}

Available tools

check_appointment

Retrieve appointment details for a given date or appointment identifier from Tekmetric.

get_shop_info

Fetch contact and basic information about a Shop stored in Tekmetric.

repair_order_status

Query the current status of a repair order by its ID.