MemPalace/mempalace
MemPalace is an open-source, local-first AI memory system that stores and retrieves conversation history verbatim using semantic search, supporting pluggable backends like ChromaDB and integrating ...
Awesome MCP › Knowledge & Memory
snexus/llm-search
snexus/llm-search is an advanced Retrieval-Augmented Generation (RAG) system designed to facilitate efficient querying and interaction with local document collections using customizable Large Language Models (LLMs). The system supports integration via an MCP (Model Context Protocol) server, enabling clients such as Cursor, Windsurf, or VSCode GH Copilot to interact seamlessly with the RAG system. The package emphasizes improvements beyond basic LLM-based RAGs, including enhanced document parsing, hybrid search capabilities, HyDE (Hypothetical Document Embeddings), chat history support, deep linking, re-ranking, and customizable embeddings. The system supports fast, incremental parsing and embedding of medium-sized document bases, tested on several gigabytes of markdown and PDF files. It includes built-in parsers for markdown (.md), PDF (.pdf), and Word documents (.docx), with additional support for other formats via the Unstructured pre-processor. Interaction with embedded documents is supported through OpenAI-compatible models, HuggingFace models, and interoperability with LiteLLM and Ollama via the OpenAI API, supporting hundreds of different models. The system generates dense embeddings stored in a vector database (ChromaDB) and supports various embedding models such as Hugging Face embeddings, sentence-transformers, instructor-based models, and OpenAI embeddings. It also supports sparse embeddings using SPLADE for hybrid search. Incremental updates to embeddings are possible without re-indexing the entire document base. Additional features include table parsing, optional image parsing, retrieve and re-rank strategies for semantic search, multi-querying inspired by RAG Fusion, and optional chat history with question contextualization. The system provides simple web interfaces, deep linking into document sections, the ability to save responses for offline analysis, and a FastAPI-based API alongside the MCP server for broad client compatibility. This makes snexus/llm-search a comprehensive and flexible RAG solution for advanced document interaction and querying using MCP.
https://github.com/snexus/llm-search
Also in Knowledge & Memory
siyuan-note/siyuan
SiYuan is a privacy-first, self-hosted personal knowledge management system featuring block-level referencing, Markdown WYSIWYG editing, and support for the Model Context Protocol (MCP).
topoteretes/cognee
Cognee is an open-source project that provides scalable and modular memory capabilities for AI agents using ECL pipelines, enabling efficient management and retrieval of past interactions and diverse data types.
vectorize-io/hindsight
Hindsight is an agent memory system designed to create smarter, learning AI agents by moving beyond simple conversation history recall to encompass a biomimetic storage architecture for long-term m...
screenpipe/screenpipe
Screenpipe is a local-first, AI-powered memory tool that records computer activity (screen, audio, accessibility tree) to provide a searchable, automated context for personal AI agents and users, i...
Olow304/memvid
Memvid is a video-based AI memory library that stores millions of text chunks in MP4 files with ultra-fast semantic search, eliminating the need for traditional databases and enabling portable, efficient AI memory management.
EverMind-AI/EverOS
EverOS is an open-source Python framework for self-evolving, portable, long-term memory for AI agents, supporting multimodal ingestion and local-first persistence with Markdown as the source of truth.
MemTensor/MemOS
MemOS is a Memory Operating System for LLMs and AI agents that unifies long-term memory management, offering a single API for structured, multi-modal memory with enterprise-grade optimizations, sig...