MCP Catalogs
Home

mirobody

by thetahealth·1,042·Score 53

A personal data platform bridging health, finance data with AI via MCP servers and Claude Agent Skills.

ai-llmhealthproductivity
192
Forks
11
Open issues
this month
Last commit
2d ago
Indexed

Overview

Mirobody is an open-source, self-hosted data platform that enables users to bridge their personal health and financial data with AI capabilities. It functions as both an MCP client and server, supporting HTTP remote access, Claude Agent Skills format (SKILL.md), and works with multiple AI systems. The platform offers deep health data integration with FHIR standards, supports 300+ wearables, and provides a sophisticated agent engine with three agent types: DeepAgent, MixAgent, and BaselineAgent for different use cases and cost/quality optimization.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Personal health analysis with data from multiple wearables and Apple Health
you:Finance management by creating custom AI tools for personal financial data
you:Building domain-specific agents that leverage personal data via MCP tools
you:What makes Mirobody different from other health data platforms?
you:Can I use Mirobody with AI platforms other than Claude?

When to choose this

Choose Mirobody for health-focused personal AI applications that need to integrate with multiple health devices and standards while maintaining user privacy.

When NOT to choose this

Don't choose Mirobody if you need general-purpose AI without health data integration, or if you prefer a simpler deployment without Docker or PostgreSQL.

Tools this server exposes

12 tools extracted from the README
  • ls

    List files in a directory

  • read_file

    Read the contents of a file

  • write_file

    Write content to a file

  • edit_file

    Edit an existing file

  • execute

    Execute shell commands in a sandboxed environment

  • glob

    Find files matching a pattern

  • grep

    Search for text patterns in files

  • query_health_data

    Query health data from various sources

  • analyze_health_trends

    Analyze health trends and patterns

  • upload_file

    Upload files for health data extraction

  • search_indicators

    Search for health indicators using medical codes

  • create_agent

    Create a custom health agent

Note: Tools were inferred from various sections of the documentation, particularly the file operations section and the health data pipeline section. The README mentions built-in tools like file ops, code execution, and health data tools, but does

Comparable tools

health-mcppersonal-data-connectorhealthkit-mcpopen-mcp

Installation

Quick Start with Docker

git clone https://github.com/thetahealth/mirobody.git
cd mirobody
./deploy.sh

This will set up a complete environment with Postgres, Redis, and Mirobody.

Use as MCP Server

Set MCP_PUBLIC_URL to expose your MCP server over HTTPS for remote access.

Claude Desktop Integration

Add to Claude Desktop configuration:

{
  "mcpServers": {
    "mirobody": {
      "command": "python",
      "args": ["-m", "mirobody.mcp.server"],
      "env": {"MCP_CONFIG_PATH": "/path/to/config"}
    }
  }
}

FAQ

What makes Mirobody different from other health data platforms?
Mirobody uniquely combines health data integration with MCP server functionality, allowing it to provide data to AI systems via standardized tools rather than just storing data.
Can I use Mirobody with AI platforms other than Claude?
Yes, Mirobody is designed as a universal tool adapter that works with ChatGPT, Claude, Cursor, and other AI systems that support MCP.

Compare mirobody with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.