Provides local semantic search and knowledge retrieval for Git repositories using vector embeddings.
Sponsored
Ndlss Memory serves as a local, intelligent memory layer for agents, designed to index files within your workspace. It leverages Qdrant for storing vector embeddings and metadata, exposing powerful semantic search and operational APIs through an mcp-server. The tool is Dockerized, offering language-specific presets for various tech stacks, and is LLM-agnostic. It features efficient delta indexing for Git repositories and a continuous watch mode, ensuring your knowledge base remains up-to-date and accessible for advanced contextual understanding and retrieval.
Key Features
01MCP-native local semantic search and knowledge retrieval
02Delta indexing for Git repositories and continuous watch mode
031 GitHub stars
04Qdrant integration for efficient vector storage and metadata management
05LLM-agnostic design for flexible integration
06Dockerized deployment with language-specific compose presets
Use Cases
01Enhancing MCP agents with a persistent, local contextual memory
02Performing semantic search within codebases to quickly find relevant information
03Accelerating development by rapidly retrieving code snippets and documentation based on natural language queries
