<img src="assets/ontoskills-banner.png" alt="OntoSkills: Neuro-Symbolic Skill Compiler" width="100%"> <h1 align="center"> <a href="https://ontoskills.sh" style="text-decoration: none; color: inherit; display: flex; align-items: center; justify-content: center; gap: 10px;"> <img src="assets/ontoskills-logo.png" alt="OntoSkills Logo Inline" height="40px" style="display: block;"> OntoSkills </a> </h1> The deterministic enterprise AI agent platform. Neuro-symbolic architecture for the Agentic Web — OntoCore • OntoMCP • OntoStore <a href="docs/overview.md">Overview</a> • <a href="docs/getting-started.md">Getting Started</a> • <a href="docs/roadmap.md">Roadmap</a> • <a href="PHILOSOPHY.md">Philosophy</a> <img src="https://img.shields.io/badge/python-3.10%2B-blue?style=for-the-badge&logo=python" alt="Python 3.10+"> <img src="https://img.shields.io/badge/node.js-18%2B-green?style=for-the-badge&logo=node.js" alt="Node.js 18+"> <img src="https://img.shields.io/badge/OWL%202-RDF%2FTurtle-green?style=for-the-badge&logo=w3c" alt="OWL 2 RDF/Turtle"> <img src="https://img.shields.io/badge/SHACL-Validation-purple?style=for-the-badge&logo=graphql" alt="SHACL Validation"> <a href="#license"> <img src="https://img.shields.io/badge/license-MIT-orange?style=for-the-badge" alt="MIT License"> </a>

What is OntoSkills?

OntoSkills transforms natural language skill definitions into validated OWL 2 ontologies — queryable knowledge graphs that enable deterministic reasoning for AI agents.

The problem: LLMs read skills probabilistically. Same query, different results. Long skill files burn tokens and confuse smaller models.

The solution: Compile skills to ontologies. Query with SPARQL. Get exact answers, every time.

flowchart LR
    CORE["OntoCore<br/>━━━━━━━━━━<br/>SKILL.md → .ttl<br/>LLM + SHACL"] -->|"compiles"| CENTER["OntoSkills<br/>━━━━━━━━━━<br/>OWL 2 Ontologies<br/>.ttl artifacts"]
    CENTER -->|"loads"| MCP["OntoMCP<br/>━━━━━━━━━━<br/>Rust SPARQL<br/>in-memory graph"]
    MCP <-->|"queries"| AGENT["AI Agent<br/>━━━━━━━━━━<br/>Deterministic<br/>reasoning"]

    style CORE fill:#e91e63,stroke:#2a2a3e,color:#f0f0f5
    style CENTER fill:#abf9cc,stroke:#2a2a3e,color:#0d0d14
    style MCP fill:#92eff4,stroke:#2a2a3e,color:#0d0d14
    style AGENT fill:#6dc9ee,stroke:#2a2a3e,color:#0d0d14

Why OntoSkills?

| Problem | Solution | |---------|----------| | LLMs interpret text differently each time | SPARQL returns exact answers | | 50+ skill files = context overflow | Query only what's needed | | No verifiable structure for relationships | OWL 2 formal semantics | | Small models can't read complex skills | Democratized intelligence via graph queries |

For 100 skills: ~500KB text scan → ~1KB query

→ Read the full philosophy

Quick Start

# Install
pip install ontoskills

# Compile skills to ontology
ontoskills init-core
ontoskills compile

# Query the knowledge graph
ontoskills query "SELECT ?skill WHERE { ?skill oc:resolvesIntent 'create_pdf' }"

Or use npx ontoskills without installing.

→ Full installation guide

Components

| Component | Language | Status | Description | |-----------|----------|--------|-------------| | OntoCore | Python | ✅ Ready | Skill compiler to OWL 2 ontology | | OntoMCP | Rust | ✅ Ready | MCP server for semantic skill discovery | | OntoStore | TBD | 📋 Planned | Versioned skill registry | | skills/ | Markdown | Input | Human-authored skill definitions | | ontoskills/ | Turtle | Output | Compiled, self-contained ontologies |

Documentation

Overview — What is OntoSkills and why it matters
Getting Started — Installation and first steps
Architecture — How the system works
Knowledge Extraction — Extracting value from skills
Registry & Packages — Package distribution and import
Roadmap — Development phases

<a id="license"></a>License

MIT License — see LICENSE for details.

Ontoskills

Install / Use

README