oml-mcp

Resolve OML ontology questions via discovery-driven retrieval and shape-validated updates.

Updated Aug 27, 2026

Search Tags:#ontology #validation #reasoning #model update #sparql #mcp tools #oml

Ready to run

npx skills add https://github.com/dwagmuse/dw-oml-3 --skill oml-mcp

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: oml-mcp
Source: https://github.com/dwagmuse/dw-oml-3/tree/main/.claude/skills/oml-mcp
Command: npx skills add https://github.com/dwagmuse/dw-oml-3 --skill oml-mcp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you work with OML ontologies and models without guessing IRIs or predicates, so you can reliably retrieve facts or make validated updates.

Core Features & Use Cases

OML Retrieval Workflow: Uses search and ontology-aware discovery (about, members/paths) before running SPARQL, reducing brittle or incorrect queries.
Shape-Guided Updates: Uses oml_shapes to ensure you assert only the exact predicates and required fields supported by the target class.
Validation-First Safety: Supports previewing oml_update plans and running oml_validate to catch lint and correctness issues before finalizing.

Quick Start

Ask an OML question such as “What members define the concept of X in this workspace?” and rely on the skill to search, explain relevant definitions, and run precise SPARQL only if needed.

Frequently Asked Questions about oml-mcp

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I query an OML ontology without knowing the exact IRIs or predicates?▼

To query an OML ontology without hardcoding IRIs, use a discovery-driven retrieval workflow that searches the workspace, explores ontology members and paths, and generates SPARQL-ready queries based on the actual model structure.

What is the safest way to create or update instances in an OML workspace?▼

The safest way to update OML instances is to use shape-guided validation, which derives exact sh:targetClass and sh:path IRIs from oml_shapes before asserting predicates, followed by running oml_validate to catch correctness issues.

How does shape-guided validation work for OML model updates?▼

Shape-guided validation works by retrieving OML shapes to ensure you assert only the exact predicates and required fields supported by the target class, preventing invalid modifications before they are finalized in the workspace.

Can I preview an OML update plan before applying changes to the model?▼

Yes, you can preview oml_update plans to review pending changes and run oml_validate to catch lint and correctness issues before finalizing any modifications to the OML model.

Why do my SPARQL queries against the OML ontology keep returning incorrect results?▼

Your SPARQL queries may return incorrect results if you guess IRIs or predicates instead of using ontology-aware discovery first; running search, about, and members/paths before querying ensures precise SPARQL execution.