Vector RAG Tool

소개

This skill provides a robust toolkit for managing local vector stores and performing semantic searches across your documentation and source code. By leveraging Ollama for embeddings and FAISS for vector search, it allows for sophisticated Retrieval-Augmented Generation (RAG) workflows where content is retrieved based on meaning rather than simple keyword matches. It is highly configurable, supporting custom chunking strategies for different file types, and can be integrated with AWS S3 for cloud-based storage, making it ideal for developers who need context-aware search capabilities within their local development environment.

주요 기능

• Configurable chunking and overlap settings to optimize search relevance for code or text
• Complete vector store management including creation, deletion, and metadata inspection
• High-performance vector search powered by FAISS for local RAG operations
• Support for S3 backend integration to store and query vectors in cloud environments
• Semantic indexing of source code and documentation using Ollama embeddings
• 0 GitHub stars

사용 사례

• Performing semantic search over a large codebase to find relevant logic or functions
• Managing multiple domain-specific vector stores across different development projects
• Building a local RAG system for internal project documentation and markdown files