Enables searching of vectorized Cursor IDE chat history using LanceDB and a local LLM.
Sponsored
Vectorizes and indexes chat history from the Cursor IDE to enable similarity searches via a self-hosted API. It extracts user prompts, generates embeddings using a local Ollama instance, and stores them in a LanceDB vector database. A FastAPI application serves as an API endpoint, allowing users to query the vectorized chat history for Retrieval Augmented Generation (RAG) and other LLM-based analysis.
Key Features
01Extracts prompts from Cursor IDE's local chat history
02Provides health check and search API endpoints
03Includes a Dockerized FastAPI application for searching
042 GitHub stars
05Generates text embeddings using a local Ollama instance (nomic-embed-text)
06Stores prompts and embeddings in a LanceDB vector database
Use Cases
01Retrieval Augmented Generation (RAG) for LLMs
02Analyzing chat history to identify patterns and improve code
03Searching past coding conversations and solutions
