Loading languages...
NC

NCBI Datasets MCP

@GooTec

No description provided.

ncbi
datasets
biology
genetics
genomics
bioinformatics

NCBI Datasets MCP Server Protocol

Model Context Protocol for the NCBI Datasets MCP server, following the MCP specification format.

Features

  • OpenAPI-driven: Loads the official NCBI Datasets v2 OpenAPI 3.0 spec for stub generation and documentation.
  • Modular endpoints: Genome, Gene, BioSample, Prokaryote, Virus, Organelle, Taxonomy, Version.
  • Containerized: Docker Compose & Kubernetes manifests included.
  • Interactive docs: Built-in Swagger UI (/docs).
  • SDK generation: Generate language-specific clients via openapi-generator-cli.

Use with Goose

Option 1: One‑click install
Copy into browser address bar to add extension:

goose://extension?cmd=npx&arg=-y&arg=@your-org/mcp-ncbi-datasets&id=ncbi-datasets-mcp&name=NCBI%20Datasets%20MCP&description=wraps%20NCBI%20Datasets%20OpenAPI%20in%20MCP%20server

Option 2: Manual (desktop or CLI)

{
  "mcpServers": {
    "ncbiDatasets": {
      "command": "npx",
      "args": ["-y", "@your-org/mcp-ncbi-datasets"]
    }
  }
}

Use with Other MCP Clients

{
  "mcpServers": {
    "ncbiDatasets": {
      "command": "npx",
      "args": ["-y", "@your-org/mcp-ncbi-datasets"]
    }
  }
}

Development

  1. Clone repo:
    git clone https://github.com/your-org/mcp-ncbi-datasets.git
    cd mcp-ncbi-datasets
    
  2. Install dependencies:
    npm install
    
  3. Run server:
    npm start
    

Installation

Smithery (Claude Desktop)

npx -y @smithery/cli install @your-org/mcp-ncbi-datasets --client claude

Manual

npm install -g @your-org/mcp-ncbi-datasets

Usage

Start the MCP server:

mcp-ncbi-datasets

Or via NPX in your MCP config:

{
  "mcpServers": {
    "ncbiDatasets": {
      "command": "npx",
      "args": ["-y", "@your-org/mcp-ncbi-datasets"]
    }
  }
}

Tools

start_server

Launches the MCP server.

{
  "tool": "start_server",
  "parameters": {}
}

get_version

Retrieve current service versions.

  • Parameters: none
{
  "tool": "get_version",
  "parameters": {}
}

get_genome_annotation_report

Fetch genome annotation report by accession.

  • Parameters:
    • accession (string, required)
{
  "tool": "get_genome_annotation_report",
  "parameters": {
    "accession": "GCF_000001405.39"
  }
}

download_genome_package

Download genome data package.

  • Parameters:
    • accessions (array of strings, required)
    • format (string, optional: fasta, jsonl, tsv)
{
  "tool": "download_genome_package",
  "parameters": {
    "accessions": ["GCF_000001405.39"],
    "format": "fasta"
  }
}

get_gene_dataset_report

Retrieve gene dataset report.

  • Parameters:
    • gene_ids (array of ints, required)
{
  "tool": "get_gene_dataset_report",
  "parameters": {
    "gene_ids": [672, 7157]
  }
}

download_gene_package

Download gene sequences and metadata.

  • Parameters:
    • gene_ids (array of ints, required)
    • include (array of strings, optional: transcript, protein)
{
  "tool": "download_gene_package",
  "parameters": {
    "gene_ids": [672],
    "include": ["transcript","protein"]
  }
}

get_biosample_report

Fetch BioSample report by accession.

  • Parameters:
    • accession (string, required)
{
  "tool": "get_biosample_report",
  "parameters": {
    "accession": "SAMN00000000"
  }
}

get_taxonomy_subtree

Retrieve filtered taxonomy subtree.

  • Parameters:
    • taxon (int, required)
    • level (string, optional)
{
  "tool": "get_taxonomy_subtree",
  "parameters": {
    "taxon": 9606,
    "level": "family"
  }
}

License

MIT

# mcpServer Config

{
  "mcpServers": {
    "ncbiDatasets": {
      "command": "npx",
      "args": [
        "-y",
        "@your-org/mcp-ncbi-datasets"
      ]
    }
  }
}

# stdio

npx -y @your-org/mcp-ncbi-datasets
Transport:
stdio
Language:
javascript
Created: 4/26/2025
Updated: 5/25/2025