Cogmem MCP Server

A self-improving, verifiable memory layer for AI coding agents. A self-improving, verifiable memory layer for AI coding agents. It runs locally over stdio via the published package.

People who work with agent memory, did and embeddings and want it reachable from Claude, Cursor, VS Code, or another MCP client. The project is written in Python.

NO LONGER MAINTAINED

LAST COMMIT 2026-09-02 · ★ 2 · VERIFIED 2026-09-18

Apache-2.0 · Python servers · how we verify → /methodology

01 · Install Cogmem

before you install - you'll need

Set COGMEM_BIN before connecting. COGMEM_HOME is optional or environment-specific per the README.

Claude Code

claude mcp add writerslogic-cogmem -- uvx cogmem

Claude Desktop / Cursor / VS Code - add to config

{
  "mcpServers": {
    "writerslogic-cogmem": {
      "command": "uvx",
      "args": [
        "cogmem"
      ]
    }
  }
}

Same JSON for Cursor. For VS Code, rename the top-level key from `mcpServers` to `servers`.

Using another client? Same JSON, different key

Claude Desktop · mcpServers

Cursor · mcpServers

VS Code · servers

Windsurf · mcpServers

Zed · context_servers

Cline · mcpServers

Roo Code · mcpServers

Continue · mcpServers

LibreChat · mcpServers

Gemini CLI · mcpServers

Codex CLI · mcp_servers

Full setup guides: every client.

02 · Evidence

Security posture

What to check before giving this server access to your agent - from the registry, GitHub, and our own probes. We don't score safety; we show what's verifiable.

runs as local process (stdio) - runs on your machine with your user's permissions

repository archived on GitHub - read-only; no future fixes, security or otherwise

license Apache-2.0 - declared in the repository

pypi package cogmem - check the name against the project README before installing (PyPI has no namespace ownership)

registry namespace io.github.writerslogic is GitHub-verified and matches the repo owner

03 · What Cogmem can do

Prose above is summarized from the project's README and registry record - no invented capabilities.

Latest releases

v2.7.1 · 2026-07-13

Adds official MCP Registry publishing on release.

v2.7.0 · 2026-07-06

cogmem is now a pip-installable package. · Install · bash · pip install cogmem # CLI + MCP server + verifiable-memory tools · pip install 'cogmem[recall]' # add local semantic recall · cogmem init…

v2.6.0 · 2026-06-29

cogmem 2.6.0 · Provenance hardening, an honest efficacy signal, and the verification surface. · Verifiable memory · Issuer pinning - credentials, the transparency log, and the signed tree head are now pinned to a TOFU…

04 · Who maintains Cogmem

cogmem is maintained by writerslogic. We track 2 MCP servers from writerslogic - 1 actively maintained, 60 combined GitHub stars, oldest repo from Sep 2025. Full record: all servers from writerslogic.

scrivener-mcp MCP server for Scrivener: read, write, analyze, and search your manuscripts with AI assistants. ★ 58

05 · Facts

repository: github.com/writerslogic/cogmem
category: memory - dead as of 2026-09-18.
release cadence: 4 releases in the last 90 days (latest 2026-07-13)
registry: io.github.writerslogic/cogmem (active, first published 2026-07-13)
packages: pypi:cogmem

06 · Cogmem FAQ

What is Cogmem?

A self-improving, verifiable memory layer for AI coding agents. A self-improving, verifiable memory layer for AI coding agents. It runs locally over stdio via the published package.

Is Cogmem still maintained?

Its last commit was 2026-09-02 (as of 2026-09-18) - we classify it as dead.

How do I install Cogmem?

Run `uvx cogmem`. The README documents 2 environment variables (COGMEM_HOME, COGMEM_BIN) to set first. Set COGMEM_BIN before connecting. COGMEM_HOME is optional or environment-specific per the README. You can also paste the ready-made client config above.

Does Cogmem run locally?

Yes - it's a stdio server: it runs on your machine (via uvx) with your user's permissions. Your data stays local unless the server itself calls external APIs.

07 · Alternatives to Cogmem

Alternatives to Cogmem

Maintained memory servers if Cogmem isn't the fit.

Pairs well with

Servers that cover what Cogmem doesn't - only shown when the pairing reason fits the companion.