ShelbyMCP is a local, Rust-based Model Context Protocol (MCP) server that provides long-term, project-aware knowledge-graph memory for AI agents. Designed to work with multiple MCP-compatible clients (such as ChatGPT, Claude, and Gemini), it offers durable, searchable, and structured memory stored in a local SQLite database, featuring graph traversal, scoped context briefing, and trust fencing. The system performs no model inference, ensuring privacy and agent-driven operations, and exposes an HTTP+OAuth interface for integration and secure access. ShelbyMCP is suitable for developers building AI and agentic applications that require rich, persistent memory and context management locally.
Visit Shelbymcp's official website for product details and getting started.
Comprehensive documentation for installing and using ShelbyMCP.
Community discussions and issue tracking for ShelbyMCP.
Code examples and usage scenarios for implementing ShelbyMCP in projects.
Comprehensive API reference and usage instructions for ShelbyMCP.
Guide on setting up and using ShelbyMCP with various clients.