MCP OKN Server

A Model Context Protocol (MCP) server providing seamless access to SPARQL endpoints with specialized support for the OKN knowledge graphs. Query 40+ scientific knowledge graphs using natural language through your favorite AI assistant.

What it does

The MCP OKN Server connects AI assistants — such as Claude Desktop, ChatGPT, VS Code Copilot, and others — directly to the FRINK-hosted knowledge graphs. Instead of writing SPARQL by hand, you describe what you need in plain English and the server translates your request into structured queries across one or more graphs.

Key features

• FRINK integration — automatic detection and documentation linking for all FRINK-hosted knowledge graphs
• Ontology-driven search expansion — queries automatically expand using ontology hierarchies (MONDO, HP, GO, UBERON)
• Federated queries — query multiple endpoints simultaneously in a single prompt
• Automatic documentation — registry links and metadata for every OKN knowledge graph
• Flexible configuration — supports both FRINK and custom third-party SPARQL endpoints

How to connect

A hosted instance of the mcp-proto-okn assistant endpoint is available at https://apps.okn.us/okn-mcp/mcp. Point your MCP-capable client at that URL — no local installation required. Visiting it directly in a browser returns a JSON-RPC error (the endpoint speaks JSON-RPC, not HTTP GET); that is expected and does not indicate the server is down. Setup steps below cover the two most common clients.

What this enables

Once connected, your assistant can discover OKN graphs, inspect graph metadata, run SPARQL queries, bridge identifiers, and return answers with source graph context. It is most useful for open-ended cross-graph questions.

Quick verification prompt

“Generate a table of all OKN knowledge graphs with two columns: KG name and description. Use only the proto-okn connector tools.”

Claude Desktop

Requires a Claude Pro or Max subscription. Download Claude Desktop from claude.ai/download.

1. Launch Claude Desktop and open Claude → Settings.
2. Choose Connectors from the Settings menu and click Add custom connector.
3. Enter the
   • Name: proto-okn
   • URL: https://apps.okn.us/okn-mcp/mcp
4. Once proto-okn appears in the connector list, click Configure and set tool permissions to Always allow.
5. Start a new chat, click the + button, toggle proto-okn on, and turn Web search off.
6. Try the verification prompt above. If the assistant returns a graph list, the connector is working.

Detailed walkthrough with screenshots: Claude Desktop setup guide.

ChatGPT

Requires a ChatGPT subscription and the ChatGPT web app — the desktop app does not currently support custom MCP services.

Note: ChatGPT may not perform as well as Claude models for this task. For best results, use Claude Sonnet or Claude Opus.

1. Sign in at chatgpt.com and click your profile name/avatar in the bottom-left corner.
2. Choose Settings from the menu, then open Apps → Advanced settings.
3. Toggle Developer mode on. MCP services only work with developer mode enabled.
4. Click Create app and enter:
   • Name: proto-okn
   • URL: https://apps.okn.us/okn-mcp/mcp
5. Start a new chat, click +, choose ... More ›, and select proto-okn. Turn Web search off.
6. Try the verification prompt above. If the assistant returns a graph list, the connector is working.

Detailed walkthrough with screenshots: ChatGPT setup guide.

Compatible clients

Any MCP-compliant application works. Popular choices include:

• Claude Desktop
• Claude Code
• ChatGPT (with subscription)
• VS Code with GitHub Copilot
• Cursor

Example queries

Once connected, try prompts like these:

• “Generate a table of all OKN Knowledge Graphs with columns for name and description”
• “@spoke-genelab: Give a high-level overview of this knowledge graph”
• “Find manufacturers of CNC machines in Ohio”

Review the full set of single-KG and cross-KG queries, along with their chat transcripts.

If you find an issue or would like to contribute additional example queries, please file an issue.