Hermes Atlas
Memory & knowledge · works with Hermes Agent

MeMesh

PCIRCLE-AI/memesh

Local SQLite memory and messaging layer for coding agents, with a native Hermes Agent memory provider

In short

MeMesh is a local collaboration layer for AI coding agents that stores shared memory in one SQLite file and lets agents exchange messages, with a native memory-provider plugin for Hermes Agent.

What MeMesh does

MeMesh keeps decisions, lessons and the links between them in a single SQLite file at ~/.memesh/knowledge-graph.db, with no Docker and no cloud required. It carries your preferences across sessions, lets agents leave each other messages through a durable inbox, and keeps a handoff of where work stopped. A decision stored from Claude Code is available to Codex or Cursor the next day.

The integrations differ in depth. Claude Code gets a plugin with hooks, MCP tools and a /memesh skill, and the README counts 10 hooks across the Claude Code and Codex integrations. Codex CLI works through a plugin or the memesh-mcp server, Gemini CLI, Cursor, Cline and other MCP clients through memesh-mcp, and your own scripts through an HTTP API started with memesh serve. Hermes Agent is listed with a native memory-provider plugin. A dashboard at http://localhost:3737/dashboard has 4 tabs and 11 languages. The README notes that automatic capture and delivery depend on the host integration.

Key features

  • One SQLite knowledge graph file, with no Docker or cloud
  • Native memory-provider plugin for Hermes Agent
  • Durable inbox for messages between agents
  • Work handoff notes for the next session
  • MCP server memesh-mcp for Codex, Gemini CLI, Cursor, Cline and others
  • Local dashboard on port 3737

When to use it

  • Sharing decisions and lessons between Hermes, Claude Code and Codex
  • Passing questions and results between agents working on one project
  • Picking up a project in a new session with a handoff briefing

Who it is for: Developers who use more than one coding agent and want one local memory and messaging layer.

How it fits with Hermes Agent

The README lists Hermes Agent with a native memory-provider plugin documented in docs/platforms/hermes-agent.md, alongside several other agents.

FAQ

What is MeMesh?

MeMesh is a local collaboration layer for AI coding agents. It stores shared memory in one SQLite file and lets agents send each other messages.

Does MeMesh work with Hermes Agent?

Yes. Hermes Agent is listed with a native memory-provider plugin, documented in the repository at docs/platforms/hermes-agent.md. Automatic capture and delivery depend on each host integration.

Is MeMesh free and open source?

Yes. The repository is released under the MIT license.

Similar memory for Hermes Agent

All memory
cx2002302-lang Zettelkasten Second Memory

Plugin that turns AI conversations into a linked Zettelkasten knowledge base with MCP tools

★ 17TypeScriptMemory okjpg Agent Context Kit

Private local-first context retrieval for Hermes agents through a read-only MCP server

★ 17PythonMemory MaiHHConnect CausaMem

Long-term causal memory system for AI agents with eight cognitive layers and intuition injection

★ 17PythonMemory ptaranat zeromem

Zero-token Rust agent memory with a Hermes Agent memory provider plugin

★ 16RustMemory rwnalds Engram

Self-hosted MCP server and dashboard for shared agent memory in a git-backed markdown vault

★ 16TypeScriptMemory amrlazw Hermes Fleet Memory

Shared vector memory for Hermes agent fleets with on-demand MCP retrieval and work/personal firewalls

★ 15PythonMemory

Related guides: SOUL.md for Hermes Agent: what it is and how to write one · Run multiple Hermes agents with profiles