# Graphora

> Graphora builds open-source tools for knowledge graphs. Our primary product is GraphCheck — semantic observability for Neo4j property graphs: declarative checks in YAML, executed read-only, returning deterministic pass/fail verdicts with evidence. We also build an open-source text-to-knowledge-graph engine and offer knowledge-graph consulting.

## GraphCheck (primary product)

GraphCheck tests a Neo4j property graph the way pytest tests code. Install: `pip install graphcheck` (Python 3.12+, Neo4j 5.x — 5 LTS or current calendar releases; Neo4j 4.4 is unsupported). It is read-only by contract: on Enterprise and Developer editions it expects a server-enforced read-only credential, and on all editions a server-side EXPLAIN preflight rejects write-capable queries. Checks are declared in YAML (conformance rules, competency questions, drift watches). Every run writes a machine-readable `results.json` (verdicts, evidence, coverage, score) and a self-contained offline `report.html`. Exit codes are a stable CI contract: 0 = all executed checks passed, 1 = error-severity finding or execution error, 2 = warnings or incomplete evaluation, 3 = the run could not be prepared or completed. Verdicts are deterministic; no LLM decides pass/fail. An optional MCP server (`graphcheck mcp serve`, install extra `graphcheck[mcp]`) exposes three agent tools: list checks, run a suite, read results.

- [GraphCheck product page](https://graphora.io/graphcheck): overview, features, and the graph quality audit
- [GraphCheck documentation](https://docs.graphora.io/graphcheck): quickstart, user guide, check reference, CI setup, troubleshooting, and the agent guide
- [GraphCheck repository](https://github.com/graphora/graphcheck): source and issue tracker
- [GraphCheck on PyPI](https://pypi.org/project/graphcheck/): releases and install
- [GraphCheck technical walkthrough](https://graphora.io/blog/graphcheck-test-your-neo4j-graph): install-to-CI tutorial with real terminal output
- [Agent-facing docs](https://docs.graphora.io/llms.txt): the docs site's own llms.txt

## Other products

- [Graphora Platform](https://graphora.io/product): open-source text-to-knowledge-graph engine (MIT) — extracts entities and relationships from unstructured text into Neo4j. Repos: https://github.com/graphora/graphora-api and https://github.com/graphora/graphora-fe
- [Compass](https://graphora.io/compass): tourism-intelligence solution built on Graphora

## Services

- [Consulting](https://graphora.io/consulting): GenAI, AI-agent, and Neo4j/knowledge-graph consulting for finance, healthcare, and consumer goods
- [Graph quality audit](https://graphora.io/graphcheck): fixed-scope, read-only audit of a production graph; the client keeps the runnable GraphCheck suite

## More

- [Blog](https://graphora.io/blog)
- [Documentation](https://docs.graphora.io)
- [Contact](https://graphora.io/contact)