Senechal

You can run a Senechal MCP Server that exposes health data from the Senechal API to your LLM applications. It provides health resources you can load into an LLM’s context, tools you can call to fetch data, and prompts to analyze health information, all through a simple MCP interface.

How to use

Load health data into your application by reading a health resource and then invoke tools to fetch the latest data or trends. Use prompts to analyze summaries or compare trends over time. You can combine resources, tools, and prompts to build a responsive health data workflow that helps an LLM reason about user health information and suggests actions.

How to install

Prerequisites: Python 3.x and a shell environment. Create a virtual environment, install dependencies, and run the MCP server.

# 1) Create a virtual environment
python -m venv venv

# 2) Activate the virtual environment
# On macOS/Linux
source venv/bin/activate
# On Windows
venv\Scripts\activate

# 3) Install dependencies
pip install -r requirements.txt

Configure your environment with the Senechal API credentials. Copy the example environment file to .env and fill in your values.

# Example values to place in .env
SENECHAL_API_KEY=your_api_key_here
SENECHAL_API_BASE_URL=https://your-api-host/api/senechal

If you are on Windows and want to run using a pre-configured MCP setup, you can use a JSON configuration that specifies the Python executable and script to start the MCP server. The environment variable SENECHAL_API_KEY is set in the config.

{
  "mcpServers": {
    "senechal-health": {
      "command": "C:\\path\\to\\venv\\Scripts\\python.exe",
      "args": [
        "C:\\path\\to\\senechal_mcp_server.py"
      ],
      "env": {
        "SENECHAL_API_KEY": "your_api_key_here"
      }
    }
  }
}

Additional setup and run commands

Start the MCP server directly from your development environment to test functionality.

python senechal_mcp_server.py

If you want to run in development mode with an inspector, you can use the MCP development command.

mcp dev senechal_mcp_server.py

To install MCP support in Claude Desktop, run this command and then select the server from the tools menu.

mcp install senechal_mcp_server.py

Configuration and usage notes

The server requires your Senechal API key and base URL to function. Set these values in the environment or provide them via the CLI configuration so that the server can fetch health data from the API.

Resources, endpoints, and data access

This MCP server exposes several resources and tools you can use to retrieve health information from the Senechal API. Use these endpoints to load data into an LLM context and to perform analyses over time.

Security considerations

Protect your API key and restrict access to the MCP server. Use secure storage for credentials and avoid logging sensitive data. Rotate keys as needed and follow your organization’s security practices when configuring environment variables.

Available tools

fetch_health_summary

Fetches a health summary for a specified period such as day, week, month, or year. You can specify metrics and span to control the amount and detail of data returned.

fetch_health_profile

Fetches the user's health profile, including demographics, medications, and supplements.

fetch_current_health

Fetches the latest health measurements, optionally filtered by measurement types.

fetch_health_trends

Fetches health trend data over a specified number of days with optional type filtering and grouping interval.

fetch_health_stats

Fetches statistical analyses of health metrics over a given period and optional type filters.