mcpbeat Sign in

codemem MCP Server

local only

codemem runs on your own machine — the client starts it, so there is no endpoint to ping. 120 installs a week from pypi. Last commit 6 Oct 2026.

Shared memory for coding agents: projects, reusable code, sessions and commits across machines

Installs per day peak 190 · avg 17 · -21% w/w
a month agotoday
120
Installs / week
pypi · codemem-mcp
2
Stars
0 open issues
6 Oct 2026
Last commit
from GitHub
Apache-2.0
License
Python

What the code does

We read the source, 5 d ago · rules ccca72095570

Capabilities

What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.

        r = subprocess.run(["git", "-C", str(p), *a], capture_output=True, text=True, timeout=30)
 try{$('#main').innerHTML=await fn(arg)}catch(err){$('#main').innerHTML=`<p class="dirty">error: ${esc(err.message)}</p>`}}

Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.

This code can reach further than it looks

We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.

Three servers free · no card

Connect this server

This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.

run in your terminal
claude mcp add codemem -- uvx codemem-mcp
~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "codemem": {
      "args": [
        "codemem-mcp"
      ],
      "command": "uvx"
    }
  }
}
~/.codex/config.toml
[mcp_servers.codemem]
command = "uvx"
args = ["codemem-mcp"]
.cursor/mcp.json
{
  "mcpServers": {
    "codemem": {
      "args": [
        "codemem-mcp"
      ],
      "command": "uvx"
    }
  }
}
.vscode/mcp.json
{
  "mcpServers": {
    "codemem": {
      "args": [
        "codemem-mcp"
      ],
      "command": "uvx"
    }
  }
}

This one needs environment variables set before it will start: CODEMEM_DATA (Directory for the database and backups), CODEMEM_DB (Path to the SQLite database (overrides CODEMEM_DATA)), CODEMEM_HOST (Address to bind. The server has no authentication; use 127.0.0.1 unless the network is trusted), CODEMEM_PORT (Port for MCP (/mcp) and the web UI. The transport URL assumes 8055), OLLAMA_URL (Ollama base URL for embeddings (semantic search). Without it, search is BM25 only), CODEMEM_EMBED (Set to false to disable embeddings entirely), GIT_ROOT (Directory of bare git repos to read for the commit feed). The author declared them in the registry entry; get the values from the project itself.

Alternatives to codemem

same job, measured the same way
Engrm
by dr12hes

Shared memory across devices, sessions, and agents for OpenClaw, Claude Code, and Codex.

32 installs/wk local only
L
logbook-mcp
by abduljabbarbxr

Project memory across sessions for AI agents. A queryable decision log in the repo.

34 installs/wk local only
TeamShared
by xhad

Shared memory for coding agents. Stop re-explaining your codebase every session.

92 tools answering
FairMind Project Context
by fairmind

Project memory for coding agents: requirements, decisions, code graph and delivery telemetry.

answering
Global Agent Memory
by ozankasikci

Shared, project-aware memory for Claude Code, Codex, and other agents with human review.

116 installs/wk local only
Codeledger
by codeledgerecf

Deterministic context selection for AI coding agents. Local memory that compounds across sessions.

1 312 installs/wk local only
Memoize — persistent agent memory
by harrisonford

Persistent cross-session memory shared by Codex, Claude Code, ChatGPT, and other AI agents.

answering
Sirro
by jonathanmberg

Shared memory for AI coding agents. Save once, reuse from Cursor, Claude Code, Codex.

answering

codemem — questions

Answers built from our own checks of this server.

Why is there no uptime for codemem?
codemem runs on your own machine over stdio — there is no network address to reach, so uptime cannot be measured for it by anyone. What can be measured is adoption: the pypi package codemem-mcp was installed 120 times last week.
How do I connect codemem?
Copy the ready config from this page — we generate it for Claude Code, Claude Desktop, Codex, Cursor and VS Code, each with the file path that client actually reads. It runs locally, so the command pulls codemem-mcp straight from pypi; nothing to host, nothing to sign up for.
How many people use codemem?
The pypi package codemem-mcp was installed 120 times in the last week. Week over week that is -21%. We show installs rather than GitHub stars on purpose: a star is a bookmark, an install is someone actually running it.
Is codemem open source?
Yes — it is published under the Apache-2.0 licence, written in Python and 2 stars on GitHub. The source link is on this page, so you can read exactly what it does with your data before you connect it.