What Makes Remembrances MCP Special?
Traditional AI agents are stateless - they forget everything between conversations. Remembrances MCP solves this by providing persistent memory, semantic search, and relationship mapping while keeping your data private and secure.
Key Features
Privacy-First Local Embeddings
Generate embeddings completely locally using GGUF models. Your data never leaves your machine, ensuring complete privacy and security.
GPU Acceleration
Take advantage of hardware acceleration with support for Metal (macOS), CUDA (NVIDIA GPUs), and ROCm (AMD GPUs).
Multiple Storage Layers
Key-Value Store for simple facts, Vector/RAG for semantic search, and Graph Database for relationship mapping.
Knowledge Base Management
Manage knowledge bases using simple Markdown files, making it easy to organize and maintain your AI’s knowledge.
Flexible Integration
Support for multiple embedding providers: GGUF Models (local), Ollama (local server), and OpenAI API (cloud-based).
Privacy Control
Keep sensitive data local with GGUF embeddings and embedded SurrealDB - no cloud dependencies required.
Why Choose Remembrances MCP?
Remembrances MCP empowers your AI agents with powerful memory capabilities while maintaining complete control over your data.
Persistent Memory
Store facts, conversations, and knowledge permanently. Your AI remembers what matters.
Semantic Search
Find relevant information using vector embeddings. Smart search that understands context.
Relationship Mapping
Understand connections between different pieces of information using graph database capabilities.
Use Cases
Personal AI Assistants
Remember user preferences and past conversations to provide a truly personalized experience.
Research Assistants
Build and query knowledge bases from documents, papers, and research materials.
Customer Support
Maintain context across multiple interactions for better customer service.
Development Tools
Store and retrieve code snippets, documentation, and technical knowledge.
AI Agents
Learn from experiences and improve the results.
Enterprise knowledge retention
The knowledge learned for the workers is saved in a company shared database.
Technology Stack
Remembrances MCP is built with modern, proven technologies:
- Language: Go 1.20+ for performance and reliability
- Database: SurrealDB (embedded or external) for flexible data storage
- Embeddings: GGUF models via llama.cpp for local, privacy-first embeddings
- Protocol: Model Context Protocol (MCP) for seamless AI integration
Open Source & Community
Remembrances MCP is open source and available on GitHub.
We welcome contributions from the community! Whether you want to report a bug, suggest a feature, or contribute code, we’d love to hear from you.