Understand any codebase in minutes.
CodeNib is a multi-view data system for serving repository context to coding agents.
One repository, coordinated views
The Wiki, Dependency Map, and Ask consume manifest-linked lexical, semantic, and structural views from one incremental compiler.
Ask, with receipts
Every answer cites the file and line it came from. Click a citation to open the source.
Compiler-grade graphs
Dependency maps come from SCIP symbol resolution, not model guesses. Edges link to definitions and references in the source.
Wikis that write themselves
Architecture pages are generated per subsystem and cached, so every page loads instantly.
From the blog
Implementation notes and measured tradeoffs in source-linked repository analysis.
Let a Model Plan grep, Then Let Jev Rank the Code
BM25, dense embeddings, hybrid retrieval, and model-planned grep on 100 CodeNib Base issues: accuracy, latency, and cost across complete retrieval and reranking paths.
Read the article →