CheMiguel23/MemoryMesh

Overview

MemoryMesh is a knowledge graph server that stores structured memory for AI models using the Model Context Protocol (MCP). It allows agents to persist and retrieve relationships between entities, enabling long-term context across interactions.

Best for

Best for
Developers building AI agents or assistants that need persistent, structured memory using the Model Context Protocol

Use cases

• Provide persistent memory for AI assistants across multiple sessions
• Store structured relationships (e.g., user preferences, task dependencies) as a knowledge graph
• Integrate with MCP-compatible clients to give agents recall of past interactions

Notes

MemoryMesh is a knowledge graph server that stores structured memory for AI models using the Model Context Protocol (MCP). It allows agents to persist and retrieve relationships between entities, enabling long-term context across interactions.

342 stars on GitHub. Last updated 2026-03-01. Licensed MIT.

Use cases

• Provide persistent memory for AI assistants across multiple sessions
• Store structured relationships (e.g., user preferences, task dependencies) as a knowledge graph
• Integrate with MCP-compatible clients to give agents recall of past interactions

Pros

• Uses the Model Context Protocol for a standardized memory interface
• Graph format captures rich, linked data beyond flat key-value stores
• Written in TypeScript, offering type-safe development and easy integration

Cons

• Requires a separate server process to run alongside your application
• Only works with clients that implement the MCP specification
• Relatively new project with a moderate community (342 stars) and limited production track record

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.