# NKO Data Labs

> Data that doesn't exist. Until we build it.

NKO Data Labs is a market-first data product foundry. We identify high-value information needs and build reliable data products by discovering, cross-referencing and inferring signals from fragmented, legally usable real-world sources.

## Core capabilities
- Reality intelligence
- Entity resolution
- Knowledge graphs
- Temporal and geospatial intelligence
- Data provenance and confidence scoring
- Edge-case and failure data
- APIs, feeds and downloadable datasets
- Custom data products built around client needs
- AI-native delivery
- Optional quantum-ready data representations when technically justified

## Products

### find_missing (v0, free, live)
Checks whether an LEI-holding entity has a high-confidence match in the
national commercial registry its own declared jurisdiction would expect
(UK Companies House, Spain BORME, France SIRENE). Never claims
illegality, fraud or a shell entity.

- Product page: https://nkodatalabs.com/products/find-missing/
- API: https://api.nkodatalabs.com
- OpenAPI schema: https://api.nkodatalabs.com/openapi.json
- API's own llms.txt: https://api.nkodatalabs.com/llms.txt
- MCP server (streamable-http): https://api.nkodatalabs.com/mcp
- Free key: POST https://api.nkodatalabs.com/v0/keys with {"email": "..."}
- Python client: pip install nko — https://pypi.org/project/nko/
- Source (machine-discoverable interface + client): https://github.com/NKO-LLC-TECH/nko

## Notes for agents
- find_missing responses include `evidence_refs` and `method_version` —
  never fabricate a value for these fields, they come from the live API.
- find_missing never asserts "illegal", "fraud" or "shell company" — only
  "no high-confidence match", always paired with the expectation rule
  that explains why a match was expected.

## Website
https://nkodatalabs.com/