MCP Catalogs
Homeowl-mcp screenshot

owl-mcp

by ai4curation·15·Score 39

OWL-MCP enables AI assistants to interact with OWL ontologies via standardized MCP protocol.

knowledge-graphdeveloper-toolsai-llm
7
Forks
2
Open issues
12 mo ago
Last commit
2d ago
Indexed

Overview

OWL-MCP is a Model-Context-Protocol server that bridges AI systems with Web Ontology Language (OWL) ontologies. It provides a thread-safe interface for performing CRUD operations on OWL files, supporting formats compatible with py-horned-owl. The server maintains an in-memory model synchronized with disk files, enabling real-time updates. It's particularly well-suited for OBO-style ontologies, providing human-readable labels for entities with standard annotation properties.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:AI-assisted ontology development and management
you:Automated annotation and classification using existing ontologies
you:Collaborative knowledge graph building with AI integration
you:What OWL formats does OWL-MCP support?
you:How does OWL-MCP synchronize with other tools like Protege?

When to choose this

Researchers and knowledge engineers working with OWL ontologies who want AI assistance for creating, modifying, and maintaining ontologies.

When NOT to choose this

If you need advanced ontology visualization features or support for other knowledge representation formats besides OWL.

Tools this server exposes

12 tools extracted from the README
  • add_axiom

    Add an OWL axiom to an ontology file

  • remove_axiom

    Remove an OWL axiom from an ontology file

  • find_axioms

    Find OWL axioms in an ontology file matching certain criteria

  • create_ontology

    Create a new OWL ontology file

  • list_ontologies

    List available ontologies in the workspace

  • get_entity_labels

    Get human-readable labels for ontology entities

  • sync_with_disk

    Force synchronization between in-memory model and disk file

  • add_observer

    Register an observer for ontology change notifications

  • remove_observer

    Remove an observer from ontology change notifications

  • export_ontology

    Export an ontology to a different format

  • validate_ontology

    Validate an ontology for consistency

  • get_configuration

    Retrieve stored configuration for ontologies

Note: Tool names and functionality were extracted from the description section that mentions 'The MCP server provides function calls for finding, adding, or removing OWL axioms' and from the features listed in the README. These tools represent th

Comparable tools

protege-mcprdfox-mcponto-mcptopbraid-mcp

Installation

Installation

  1. Install Goose (Desktop or CLI version) from [goose installation](https://block.github.io/goose/docs/getting-started/installation/)
  1. Set up an LLM provider (Anthropic recommended)
  1. Install OWL-MCP extension in Goose:

- Direct install: [Install OWL-MCP](goose://extension?cmd=uvx&arg=owl-mcp&id=owl_mcp&name=OWL%20MCP) - Manual: Add uvx owl-mcp to the Extension section of Goose

  1. Start using the extension to create and manage OWL ontologies

Claude Desktop Configuration

{
  "mcpServers": {
    "owl-mcp": {
      "command": "uvx",
      "args": ["owl-mcp"]
    }
  }
}

FAQ

What OWL formats does OWL-MCP support?
OWL-MCP accepts any format supported by py-horned-owl. It follows OBO guidelines and recommends functional syntax for source files.
How does OWL-MCP synchronize with other tools like Protege?
The server keeps an in-memory model synchronized with disk files. When changes are made, they're written to disk. If you have Protege running with the same file open, Protege will automatically detect and display these updates.

Compare owl-mcp with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.