The ontology platform
for AI agents.

Design semantic contracts in Studio. Publish them to Hub. Serve them to any agent over MCP. Three surfaces, one source of truth — so every agent understands your data and delivers.

Browse the Hub Start building free

For modelers

Design ontologies visually.

Drag-and-drop classes and relationships. SHACL validation in real time. AI Copilot when you want it. Export to OWL/RDF, JSON-LD, Cypher, or publish straight to Hub.

Explore Studio →

For agent builders

Connect agents in one line.

Every Hub ontology has a live MCP endpoint. Paste the URL into Claude Desktop, Cursor, or LangChain — your agent gains typed entities, properties, and relationships your domain experts have already validated.

See how Hub works →

An agent contract, not just an ontology.

OntoBoom doesn't just model your domain. It bundles your ontology, your database mapping, and an AI-ready tool schema into a versioned package agents can consume directly — no more hand-writing SQL prompts or guessing column names.

1
Connect your database
Point OntoBoom at your PostgreSQL, MySQL, or SQL Server. We capture a schema snapshot — tables, columns, types, foreign keys.
2
Map schema to ontology
Auto-map with deterministic matching or AI suggestions. Tune rules manually — link tables and columns to classes, data properties, and relationships.
3
Export & power your agents
Download the OPS package. Drop tool-schema.json into LangChain or OpenAI function-calling. Agents query your DB with semantic concepts and the real column names, not guesses.

What's in the OPS package

ontology.ttl
OWL ontology in Turtle — classes, properties, relationships, constraints.
mapping.json
DB-to-ontology rules. Each rule maps a table or column to a semantic concept with confidence + reasoning.
tool-schema.json
AI-ready tool definition — drop into LangChain, OpenAI function calling, or any agent framework.
shapes.ttl + ops.json
SHACL validation shapes and a manifest with SHA-256 checksums for integrity verification.

LangGraph (Python) — works with any agent framework

from langgraph.prebuilt import create_react_agent
import json

with open("tool-schema.json") as f:
    tool_schema = json.load(f)

# semantic_query is defined in the OPS README — translates a
# natural-language question to SQL using mapping.json's rules
agent = create_react_agent(llm, tools=[semantic_query])

agent.invoke({"messages": [(
    "user",
    "Show investment accounts opened in the last 30 days"
)]})

OpenAI function calling

with open("tool-schema.json") as f:
    schema = json.load(f)

functions = [{
    "name": t["name"],
    "description": t["description"],
    "parameters": t["parameters"],
} for t in schema["tools"]]

OpenAI().chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user",
               "content": "Find customers with active contracts"}],
    functions=functions,
)

Full integration docs with copy-paste examples →