OLS

You can query up-to-date ontologies through a dedicated MCP server that exposes the Ontology Lookup Service (OLS) API. This server lets you search terms, explore ontologies, and navigate term hierarchies from numerous biological and medical ontologies, enabling reliable integration with AI assistants and workflow automation.

How to use

Use an MCP client to connect to the OLS MCP Server and perform a set of common actions. You can search terms across ontologies with flexible filtering, discover available ontologies and their metadata, retrieve detailed ontology or term information, explore direct term children and ancestors, and find semantically similar terms using embeddings. Typical usage patterns include asking for definitions or relationships (for example, terms related to a biological process, or the ancestry of a specific term) and then drilling down into term details to support your AI-assisted reasoning.

How to install

# Prerequisites
- Python 3.12 or higher
- uv package manager

# Install uv if needed (macOS/Linux)
curl -LsSf https://astral.sh/uv/install.sh | sh

# Install the OLS MCP Server
# Clone the repository
git clone https://github.com/seandavi/ols-mcp-server.git
cd ols-mcp-server

# Install dependencies
uv sync

# Optional: install as a tool
uv tool install .

# Alternative quick start (if you have a ready-made package)
ols-mcp-server --help

Additional content

Configuration, troubleshooting, and development guidance are provided below to help you run and maintain the OLS MCP Server in your environment.

Configuration and usage notes

To use the OLS MCP Server with Claude Desktop, add an MCP server configuration that points to the local runtime. You can run the MCP server via the uv tool or invoke it directly if installed as a standalone tool.

Troubleshooting

Common issues include ensuring the server path is correct in your client configuration, making sure the server script is executable, verifying network connectivity, and confirming you’re using a compatible Python version (3.12 or higher). If you run into permission errors, verify file permissions on the server executable. For network-related problems, check firewall settings and endpoint reachability.

API and endpoints

The server interfaces with the EBI Ontology Lookup Service API v2. Key endpoints include search, ontologies, terms, and hierarchies. These endpoints enable you to perform comprehensive ontology lookups, term lookups, and hierarchical traversals needed for ontology-enabled AI workflows.

Development setup

If you are developing or extending the MCP server, follow these steps to set up your environment and run tests.

Adding new features and tools

To add new tools, implement them in the server logic and expose them through the MCP framework. Create models for structured responses and update tests to cover new functionality. Keep documentation in sync with your changes.

API documentation excerpts

The server exposes endpoints and actions corresponding to ontology search, ontology information, term information, term children, term ancestors, and similar terms, enabling rich programmatic access to ontology data.

Notes on available ontologies

The server works with ontologies accessible through the Ontology Lookup Service, including GO (Gene Ontology), HP (Human Phenotype Ontology), MONDO, ChEBI, UBERON, and many others.

Security and environment

Environment and deployment considerations should ensure secure access to the MCP server and proper handling of API keys or embedding features if used. Use separate environments for development and production and monitor network activity to detect unusual usage.

Notes on tools and integrations

Tools and endpoints cover: Search Terms, Search Ontologies, Get Ontology Information, Get Term Information, Get Term Children, Get Term Ancestors, and Find Similar Terms. These tools enable end-to-end ontology lookup and reasoning within AI-assisted workflows.

How to configure Claude Desktop (example)

{
  "mcpServers": {
    "ols-mcp-server": {
      "command": "uv",
      "args": [
        "tool",
        "run",
        "ols-mcp-server"
      ],
      "env": {}
    }
  }
}

Available tools

Search Terms

Search for terms across ontologies with flexible filtering.

Search Ontologies

Discover available ontologies and their metadata.

Get Ontology Information

Retrieve detailed information about a specific ontology.

Get Term Information

Retrieve comprehensive details about a specific term.

Get Term Children

Find direct child terms within ontological hierarchies.

Get Term Ancestors

Retrieve parent terms and ancestors.

Find Similar Terms

Discover semantically similar terms using embeddings.