mex CLI
Shared memory for a team and its agents: it lives right in the repository and syncs through Git
- 1,781 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Writes memory files into the repository and commits them
- Connected agents may reach the network
Install
- Works in
Manual install
npx mex-agent@0.8.2 setupSets up memory in the current repository.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Writes memory files into the repository and commits them
- Connected agents may reach the network
This is third-party code. Review the repository files before installing.
What it does
mex keeps the working memory of a team and its agents inside the repository, so it is shared through plain Git rather than a separate service. The installer places mex skills into the agents' directories and sets up structure for context, patterns, specs and topics. There are commands to rebuild the graph and the wiki index, plus a local hub to browse it. The tool separates the shared part, which is committed, from the local part, which stays with you.
Who it is for. For development teams that want one project memory shared by people and agents.
Good fit when
- You need agents to remember project decisions and conventions
- You want to share team memory through Git
- You need a shared base of project context and patterns
Not a fit when
- A single person runs the project and shared memory is not needed
- You cannot commit working notes to the repository
Example request
Save to mex that we agreed to use this migration approach and add it to the project memoryLimitations
Part of the memory is committed to the repository, so watch what becomes shared. Individual connected agents have their own install and network requirements, described in the README.
How to disable. Remove the mex-agent package via your package manager and delete the .mex directories and mex skills from the repository.
FAQ
Where is the memory stored?
In the repository itself: the shared part is committed and shared via Git, the local part stays with you.
Do I need a separate service?
No, syncing goes through Git; a local hub is available for browsing.
Discussion
No questions or reviews yet.
Related
CLI toolsLocal shared project memory for coding agents: facts, decisions and commits available through MCP, CLI and hooks in any agent
A persistent-memory MCP server for coding agents: a graph of a project's concepts, architecture and decisions that survives across sessions
An MCP layer of persistent collective memory for agents: structured storage, semantic search, a relation graph and a supersession lineage
Git-native persistent project memory for agents based on Open Knowledge Format: a single Go binary, an embedded MCP server and BM25 search
An MCP server in one binary: AI reads and edits files across several projects and runs your commands
Work modes for AI coding agents: a light solo mode for daily tasks and /man with a plan, checks and review for big changes
README in short
The README describes mex as shared memory for engineers and their agents that lives in the repository and is shared through Git. Installation uses npx mex-agent with a setup command, with CLI and agent-memory modes and a global install. It lists commands to rebuild the graph and the wiki index and to run a local hub. It shows the split between shared and local memory.