For agents & integrators

Documentation

Evergreen setup guides and reference pages. Each HTML doc has a markdown twin (append .md) for LLM ingestion, indexed from llms.txt, with structured data and a GEO strategy for AI answer engines.

• MCP serverConnect Cursor, Claude, Windsurf, and other agents to the GraphCanon knowledge graph over Streamable HTTP.
• REST API & markdown twinsOpen JSON API, OpenAPI schema, and .md twins for every page - no auth required.
• Trust methodologyHow maintenance, provenance, and dependency scan signals are computed and what we never claim.
• GlossaryPlain-language definitions for AI, knowledge-graph, and data-pipeline terms - corpus, RAG, embeddings, MCP, and more.
• llms.txt indexCompact link-first index for LLM crawlers - part of our [GEO](/glossary/discovery-for-agents/generative-engine-optimization) discovery layer.
• ARD ai-catalog.json[Agentic Resource Discovery](/glossary/discovery-for-agents/ai-catalog) manifest at /.well-known/ai-catalog.json.

Quick links: MCP server, REST API, llms.txt.