ContextWeaver

ContextWeaver evolves beyond a mere semantic code retriever, establishing itself as a holistic context infrastructure for AI coding assistants. It intelligently bridges the gap between raw code and AI comprehension by employing hybrid retrieval methods, including vector, FTS, RRF fusion, and reranking to pinpoint relevant code segments. The system integrates advanced project memory capabilities to store feature definitions, architectural decisions, and project profiles, complemented by a long-term memory for non-code information. Furthermore, its cross-project hub facilitates knowledge sharing and dependency analysis across multiple repositories, all supported by an asynchronous indexing pipeline, robust observability, and optimization features to ensure retrieval reliability and performance.

Key Features

Use Cases