Build on Cogitave Core
Learn the harmonic core that every Cogitave product and agent shares: the Cogitave Core node and edge model, Cogitave Query, the native MCP interface, and how products project their state into a single queryable graph. By the end you'll be able to read the estate the same way humans and agents do. Day 0: this path is scaffolded - the module bodies are authored in a later content phase.
Modules4
Duration99 min
Levelintermediate
Prerequisites
- Basic familiarity with graph data models and APIs.
- Completion of the "Contributor onboarding" path is recommended.
Modules in this path
- Core model fundamentals Explain why Cogitave Core is one canonical property graph that humans and agents both query, what its node/edge model and identity scheme actually are, and why ADR-0001 chose a labeled property graph as the substrate.
- Query the estate with Cogitave Query Learn how Cogitave Query resolves a request against the canonical graph - lexical BM25 and dense HNSW vectors fused by Reciprocal Rank Fusion, a graph-aware rerank and ranking signals, and the bounded, read-only query_graph profile agents call directly.
- The native MCP interface to Core Read Core's MCP interface as the source of truth for how an agent queries the estate - why the surface is native rather than adapted, the protocol contract it commits to, and the tools and resources you call.
- Project products into Core See how content projects into Cogitave Core as typed, content-addressed nodes rather than separate silos, and how the fact registry's cite-not-restate rule and fact-drift scanner keep one owner per fact across the estate.