Codebase Knowledge Graph Skill

What is it

Codebase Knowledge Graph Skill is an agent skill that converts codebases, docs, SQL schemas and PDFs into a queryable knowledge graph. Rather than relying on a vector store, it uses deterministic AST parsing — positioning it as a new retrieval paradigm beyond RAG. It sits in the DevTools category and was first seen on 2026-10-13.

Why now

The term is at a nascent stage with a score of 41/100, based on a single mention across one source, github. That volume is minimal, so this is an early signal rather than an established trend — one data point, not momentum. Its interest lies in the concept itself: deterministic AST parsing as an alternative to vector-store retrieval for agent skills, which is worth watching precisely because so little has been written about it yet.

Who should care

Indie developers building AI coding agents or developer tooling should track this, since the AST-parsing approach targets a known weakness of vector-based retrieval over structured code. SaaS founders working on code search, documentation, or internal knowledge tools may find the "queryable knowledge graph" framing relevant to their roadmaps. Given only 1 mention on github and a nascent stage, this is a watchlist item for product people scanning early DevTools signals — not yet a proven direction to build on.

(Note: all figures — category, first-seen date, score, stage, source, and mention count — are drawn directly from the provided data.)