Bio MCP Server
Access PubMed, ClinVar, and OncoKB data through a unified biomedical query grammar.
Data last scanned yesterday · Reviewed 3 weeks ago
Overview
BioMCP provides a single binary interface to interact with approximately 30 biomedical data sources, including PubMed, ClinicalTrials.gov, OncoKB, and Reactome. By using a consistent command grammar, it enables users to search for genes, variants, drugs, or diseases and pivot between different data contexts without needing to manage multiple API integrations or manual identifier reconciliation. Designed for clinicians and researchers, the server supports local study analytics through command-line tools that generate charts from cBioPortal-style datasets. It also includes advanced capabilities such as g:Profiler enrichment and citation mapping, allowing for the systematic exploration of literature and biological entities.
Our verdict
With a security score of 48/100, this tool is categorized as high risk. The security architecture, specifically the authentication and read-only mode, has not yet been editorially reviewed. Given that this is a community-maintained project and not provided by an official vendor, users should exercise caution when integrating it into sensitive research environments where data isolation is critical.
- Unified access to 30 trusted biomedical sources
- Zero direct dependencies reduces supply-chain surface area
- Active development with daily commit frequency
- Supports local analysis of cBioPortal-style datasets
- Lacks external editorial or security review
- Authentication and read-only mode capabilities remain unreviewed
- Community-maintained rather than supported by an official vendor
Tools
| Tool | Description | Risk |
|---|---|---|
| search | Search for entities like genes, variants, articles, trials, etc. across various biological data sources. | low |
| skill list | List available guided investigation workflows. | low |
| discover | Perform concept resolution for biological entities. | low |
| get | Retrieve focused details for a specific entity ID, with optional section filtering. | low |
| enrich | Perform gene-set enrichment analysis. | low |
| batch | Execute multiple focused get calls in parallel. | low |
Compatibility
| Client | Local | Docker | Remote | Read-only |
|---|---|---|---|---|
| ChatGPT | ||||
| Claude Desktop | ||||
| Cursor | ||||
| VS Code | ||||
| Windsurf |
Frequently asked questions
›What is the correct PyPI package name for installation?
You must install biomcp-cli. The biomcp package on PyPI is unrelated to this project.
›Does the installer modify my shell startup files?
No, the installer does not edit your shell startup files. If the installation directory is not in your PATH, it simply prints the command needed to add it.
›Can I use BioMCP without a local installation?
Yes, you can use the Docker image ghcr.io/genomoncology/biomcp to run CLI checks or use the server via stdio.
›How do I configure BioMCP for use in Claude Desktop?
You can add it to your configuration file by setting the command to biomcp and the arguments to ["serve"] under the mcpServers key.
›Is it possible to run BioMCP as a remote server?
Yes, you can start a remote HTTP server by running the biomcp serve-http command, which allows remote clients to connect via http://127.0.0.1:8080/mcp.
›What is the purpose of the 'skill' commands?
The skill commands provide access to a catalog of worked examples and workflows, which can be installed into your agent directory to assist with guided investigations.
Alternatives
Changelog
Badge
Maintain this server? Add the live badge to your README.