ContextWeaverWeave codebase context for AI agents

Semantic code retrieval engine designed for AI coding agents — delivering precise, context-rich code snippets through hybrid search, graph expansion, and intelligent packing

Get Started →

View on GitHub

🔍

Hybrid Search

Combine vector similarity and full-text search with RRF fusion algorithm, delivering more accurate and stable code retrieval results.

🌳

AST-Aware Chunking

Leverage Tree-sitter to parse code structure, maintaining semantic integrity by chunking at function, class, and module boundaries.

🕸️

Graph Expansion

Automatically expand context from matched chunks to neighboring code, breadcrumbs, and import dependencies for complete understanding.

📦

Token-Aware Packing

Intelligently merge, rank, and trim code snippets within your token budget, ensuring LLMs receive optimal context without overflow.

🚀

MCP Integration

Built-in Model Context Protocol server provides seamless integration with Claude and other MCP-compatible AI agents.

⚡

Lightning Fast

Why ContextWeaver?

ContextWeaver is a semantic retrieval engine built for AI coding agents. It indexes codebases into searchable semantic chunks, expands context around matches, and packs the final result within a token budget so models receive more complete and relevant code evidence.

Use cases

Provide codebase retrieval tools to MCP clients
Locate functions, classes, references, and related files in large projects
Build local RAG and code context engines
Give AI agents on-demand, explainable, and controllable code context