What problem it solves

Agents can retrieve documents and query databases, but unreliable results undermine agentic analytics. Teams need a context layer that records where data lives, what its definitions mean, whether it is usable, how it should be queried, its provenance, and related concepts.

Key features

• Enterprise ontology generation — Generate semantic ontologies from database metadata, infer table relationships, and map business terms to source columns. Existing ontologies and semantic models can also be connected.

  Source: README

• Certified query generation — Ground questions in reviewed terms, relationships, and SQL attributes. Generated queries are checked against certified definitions and repaired when database validation fails.

  Source: README

• Multiple agent interfaces — Expose ontology context through MCP, CLI, and API integrations. The MCP server provides read-only tools for readiness, term discovery, answerability checks, and questions.

  Source: README

• Explainable answers and feedback — Return the generated SQL and the steps used to produce each answer, including row and chart views in Chat. Certified corrections and usage can be folded back into the ontology.

  Source: README

Quick start

Copied verbatim from the project's README.

   git clone https://github.com/NVIDIA/auto-ontology.git && cd auto-ontology
   cp .env.example .env

   docker compose up -d --wait postgres
   git clone https://github.com/NVIDIA/auto-ontology-eval.git ../auto-ontology-eval
   (cd ../auto-ontology-eval && uv sync && cp .env.example .env && \
     POSTGRES_PASSWORD='<the same password>' \
     uv run python scripts/seed/seed_wwi.py --database-name wideworldimporters --drop)

   docker compose up -d --build

Who it may suit

This may suit developers and data teams building grounded analytics, semantic catalogs, or agent workflows over enterprise databases.

Topics to discuss

• How should organizations certify business terms, relationships, and SQL attributes for agent use?

• Which provenance details should be included with every generated answer to support effective review?

• How can feedback and usage signals improve an ontology without introducing incorrect definitions?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/NVIDIA/auto-ontology#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.