Store, consolidate, and recall coding agent memories with provenance tracking using SQLite and FAISS for fast, structured knowledge access.

Ask plain-English questions about your projects — source code, infrastructure configs, runbooks, documentation, anything. glean indexes your files locally using Ollama embeddings and ChromaDB, then answers with cited file and line references. No data leaves your machine.

openai vector store automation

Local Retrieval-Augmented Generation (RAG) pipeline for document QA. Built with LangChain, Llama2 (via Ollama), and an in-memory DocArray vector store, featuring an interactive Streamlit interface for semantic PDF querying.