cogito-mem (Cogi)

Name: cogito-mem (Cogi)
Author: meridian21lab

Local

meridian21lab.github.ioMemory & Knowledge0Updated 1w ago

Repository Visit website

About cogito-mem (Cogi)

cogito-mem is a local long-term memory server for Claude Code and other agents. It preserves conversation text and selected activity on the user's machine, then retrieves relevant passages without cloud storage or AI calls to save messages.

Key features

01Stores conversations word for word
02Records commands and touched files
03Local SQLite storage
04Retrieves relevant memories per prompt
05Redacts secrets and supports capture pause

Use cases

01Recall decisions from earlier sessions
02Recover details outside the context window
03Keep agent memory on a local machine

MCP support

The local `cogi` MCP server exposes `memory_query` and `memory_append` tools to Claude.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config

{
  "mcpServers": {
    "meridian21lab-cogito-mem": {
      "command": "uvx",
      "args": [
        "cogito-mem"
      ],
      "env": {
        "COGITO_MEM_NAMESPACE": "<COGITO_MEM_NAMESPACE>"
      }
    }
  }
}

Run commanduvx cogito-mem

Environment variables to set

COGITO_MEM_NAMESPACEThe memory's name (default: <your username>-hollow).

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages

pypicogito-mem0.2.1stdio

Tools

Saved tool list · updated 6d ago

Parameter	Type	Required	Description
text	string	Yes	The note, word for word.
title	string	No	Optional short title, e.g. the project or topic.

Details

From the project README

cogito-mem

Long-term memory for Claude Code, on your own machine. Cogi the squirrel keeps every message you and Claude exchange, and brings back what Claude forgot. No API key, no cloud, no AI calls to store.

Cogi the squirrel: Claude forgets. Cogi doesn't.

Why try it

Claude stops forgetting. Every message and reply is kept word for word, with its date. When something has dropped out of the context window, or happened last week, the relevant part comes back with your next prompt.
It remembers what you did, too. The commands you ran (and whether they failed) and the files you touched, without their output.
Private and free. One folder on your computer. No account, no API key, no AI calls to save a message. Nothing leaves your machine.
Tested. On LongMemEval-S, the past session holding the answer is among the first five Cogi finds for 99.0% of 500 questions, with no AI model involved; other local memory plugins published 95.2–98.0% as of October 2026. On LoCoMo, the 1,000 tokens Cogi hands back answer 74.7% of questions right, level with rereading the whole conversation (76.5%). A search over 10 million tokens takes 26 ms on a laptop CPU. See the results.

Install

You need Claude Code and uv, on macOS or Linux (Windows is experimental). Then, in a terminal:

claude plugin install cogito-mem --marketplace meridian21lab/cogito-mem

The first start downloads the model (310 MB, about a minute). To get updates, turn on auto-update once: /plugin → Marketplaces → cogito-mem → Enable auto-update.

Step-by-step guide · Codex, Claude Desktop, Docker

Try it

Tell Claude: Let's call the staging server "acorn".
Clear its short-term memory: /clear
Ask: What did we call the staging server? Claude answers "acorn": Cogi brought it back.

What Cogi runs, downloads and stores

Runs: four Claude Code hooks and one MCP server, cogi. Each runs the released cogito-mem package from PyPI at the exact version the plugin pins (uvx cogito-mem==<version>); the plugin's uv.lock lists that release and numpy with their checksums. After each reply and at the end of a session, the hooks save the new messages from the session's transcript. At the start of a session and with each prompt, they add the few most relevant memories (at most 300 tokens per prompt). The MCP server gives Claude two tools, memory_query and memory_append, answered by a small service that listens on your computer only (127.0.0.1:7327).
Downloads, once, over HTTPS: cogito-mem and its one dependency, numpy, from PyPI (by uv); llama.cpp's llama-server from llama.cpp's official GitHub releases (skipped if you already have it); and the EmbeddingGemma 2 model (310 MB) from Hugging Face. After that, Cogi makes no network calls.
Stores: everything in ~/.cogito-mem/ on your computer: your messages, and a short note of what each turn did, in one SQLite file. API keys, tokens, private keys and passwords are replaced with [redacted] before anything is saved; tool output and file contents are never saved. Text inside <private>…</private> is skipped, and COGITO_MEM_CAPTURE=off pauses saving.
Sends: nothing. No account, no analytics, no telemetry. (The optional benchmark command, cogito-mem eval locomo-qa, which only you would run, sends benchmark questions to the LLM API you set up for it, with your own key.)
Removes cleanly: claude plugin uninstall cogito-mem@cogito-mem, then delete ~/.cogito-mem.

Every file, download and connection, with links to the source: What you install.

How it compares

	cogito-mem	claude-mem	Mem0 plugin	cognee plugin	Supermemory plugin
AI calls to store a message	none	an observer model per tool call (Haiku on your Claude plan by default)	extraction on Mem0's servers	extraction with your LLM key or `claude -p`	in Supermemory's service
Account or API key	none	optional sign-in	Mem0 Platform API key	an LLM key or the Claude CLI	API key or login
Where your memory lives	your computer	your computer (cloud sync optional)	Mem0's cloud	your computer (cloud optional)	Supermemory's cloud (local mode available)
What comes back	your exact words, with dates	AI-written observations	extracted memories	graph facts and summaries	extracted memories
Searchable after	the same turn	the observer's run	a batch (every 5 exchanges)	the next graph run	the background save

Keep your project's rules in CLAUDE.md, and let Cogi remember what happened.

Learn more


How it works	Collect, keep, find: what Cogi stores and how it finds it
Results · Benchmarks	The numbers, and how to reproduce them
API & MCP · Reference	HTTP, Python, terminal, settings, FAQ, roadmap
What you install · Privacy · Terms	Every file, download and connection; privacy and terms
Development	Running the tests and the evaluations

Early access. v0.2 is new, built and tested (over 200 tests), and still being measured. Feedback and issues are very welcome.

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

Read the getting-started guide

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.