Memory, context & research
Graphify-Labs/graphify
Navigable code knowledge graph, no vector store
- 108k
- Apache-2.0
- Python
What it is and why it matters
Turns a codebase, with its docs, SQL schemas and PDFs, into a queryable knowledge graph via local deterministic AST parsing. It is not a RAG system: there is no vector index or embeddings, and every edge is explained. It ships as a skill (/graphify) for Claude Code, Cursor, Codex and Gemini CLI.
Key points
- Local deterministic AST parsing, no embeddings or vector store
- Ingests code, docs, SQL schemas, configs and PDFs
- Skill for Claude Code, Cursor, Codex and Gemini CLI
Where we saw it
Chase AI 2026-07-08 Presented as the memory piece for large codebases: it builds a graph the agent can traverse, sitting between Obsidian and a full RAG. The host stresses it uses no vector index or embeddings. 5 Open Source Repos That Fix 95% of Claude Code's Problems The Next New Thing 2026-07-24 The guest explains it analyzes the project upfront and links docs to the app's parts, so the agent need not figure everything out from scratch on each change; he had not tested it deeply but calls the concept sound. Free ai video creation + OCR full books + private photos + more Github finds
Same category