mcpField
🧠
🧠 OSDR MCP Server + Agents

🧠 OSDR MCP Server + Agents

Local
github.com/alwaltScience & Education0Updated 80w ago
Repository

About 🧠 OSDR MCP Server + Agents

A repository of MCP server and agent examples for retrieving, analyzing, and summarizing NASA Open Science Data Repository data. It includes custom OSDR tools and workflows ranging from a simple summary to multi-agent RNA analysis and reporting.

Key features

  1. 01Fetch OSDR dataset metadata
  2. 02Filter studies by organism
  3. 03Download RNA count data
  4. 04Create a bar plot and Markdown report
  5. 05Configure LLM backends and MCP connections

Use cases

  1. 01Retrieving metadata for an OSDR dataset
  2. 02Finding studies related to an organism
  3. 03Analyzing RNA counts and producing a report

MCP support

The custom MCP server exposes tools for OSDR metadata and data workflows, with additional RNA analysis tools. Example agents also connect to official fetch and filesystem MCP servers.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

🧠 OSDR MCP Server + Agents

This repo demonstrates how to integrate MCP agents with custom tools for interacting with NASA’s Open Science Data Repository (OSDR). It includes working examples of agent workflows that fetch, analyze, and summarize biological data using the Model Context Protocol (MCP).

📦 What’s Inside

osdr_mcp/

Custom MCP server exposing tools for interacting with OSDR data:

  • osdr_fetch_metadata: Fetches metadata for a given OSDR dataset
  • osdr_find_by_organism: Filters studies by organism
  • Additional tools (e.g. RNA analysis) live in osdr_viz_tools
first_example/

A simple agent that uses two official MCP servers:

  • mcp-server-fetch (headless browser)
  • mcp-server-filesystem

This agent gathers information and generates a tweet-sized summary.

second_example/

Similar to first_example but connects to a custom MCP server defined in osdr_mcp/main_simple.py. Demonstrates how to plug in domain-specific tools like OSDR metadata fetchers.

third_example/

A full multi-agent workflow:

  1. Fetch Agent – Grabs OSD study metadata
  2. Quant Analysis Agent – Downloads RNA count data and creates a bar plot
  3. Summary Writer Agent – Generates a markdown report summarizing the analysis

Uses custom MCP servers:

  • osdr_data_fetch
  • osdr_viz_tools

⚙️ Configuration

The cp_agent.config.yaml file controls:

  • LLM backend (e.g. Ollama, OpenAI)
  • MCP server connections
  • Tool availability
  • System prompts / metadata

🚀 Getting Started

  1. Clone this repo
  2. Install dependencies: 
    pip install -r requirements.txt
  3. Run an example: 
    python first_example/main.py

Or launch the MCP server directly:

python osdr_mcp/main_simple.py

🧩 Integration Notes

This architecture is built for flexibility. You can toggle between document Q&A, RAG search, or custom analysis tools. A mode switch or UI toggle is ideal for user-facing integration. Support for Milvus-based RAG via MCP is on the roadmap.

📚 Resources

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.