I Built Context+ AST and Embeddings for Codebase Understanding
AST + embeddings for codebase searchβbut Sourcegraph Cody, Cursor, and Continue already solve this.
A super light-weight embedded code search engine CLI (AST based) that just works - saves 70% token and improves speed for coding agent π Star if you like it!
AST-based code indexing saves tokens but Cursor and Continue already do this.
Developers using AI coding agents (Cursor, Claude, etc.)
Cursor Β· Continue Β· Sourcegraph Cody
Works with Claude, Codex, OpenCode and any coding agents, I'd love to learn from your feedback!
AST + embeddings for codebase searchβbut Sourcegraph Cody, Cursor, and Continue already solve this.
Honest benchmark shows RAG overhead on trivial queries; 63% token savings on complex tasks.
49 MCP tools and 94% token reduction, but 'chat with codebase' is a crowded category.
SQLite-vec + tree-sitter chunking cuts Claude Code costs by 39% versus grep-based search.
Surfaces architectural decisions from git history β Cursor and Continue don't do this.
Tree-sitter + FTS5 + MCP = tokens saved for AI agents to actually code, not search.