THE MEMORY MCP · MODEL CONTEXT PROTOCOL · OPEN BETAOPEN SOURCE

Tired of re-explaining yourself to your agent?

MAGNEMO is Persistent Memory for AI Agents.

Kept on your computer, in plain files.

Your AI writes down what it learns.

You approve what matters.

Give your agent a notebook. Install in one paste.

Paste this to your agent. Claude Code, Cursor, Claude Desktop.

WATCH THE 60-SECOND DEMO

Your Memory lives on your computer, in plain files you can open. No network calls of its own. Add a backup remote and it pushes there, nowhere else.

Plain files — read your own Memory Open source

Works with any AI that speaks MCP. One paste for Claude Code; a config block for the rest. Claude Code, Cursor, Claude Desktop, and the rest.

Watch it grow a Memory.

A real terminal, real machine time: the install, the vault, the mount — then the session is killed, and a brand-new one resumes from Memory, quoting the receipt it resumed from.

You’ll know it worked.

Your Agent has a brain now.
→
Ask it what it remembers.
→
Close the session. Ask again.
SESSIONS DIE. THE MEMORY DOESN’T.

The case, in fifteen seconds.

WHAT IT IS
A Memory your AI keeps. It wakes up knowing what you decided, remembers what mattered, and hands off clean when the session ends.
WHY IT EXISTS
Right now your AI forgets everything when the session ends. Or it remembers wrong. Anything can get in. Nothing says who wrote it.
WHY THIS ONE
In 1969 a computer flew to the Moon with its memory woven in by hand. Cut the power and it still knew the way home. That’s Magnetic Memory. It stays. Every wire was placed by a person and checked. That’s Governed Memory. It’s true. MAGNEMO is both.

Every Agent needs a Memory; the only question is whether it’s one you can trust.

Any memory works on day one. Watch what happens by day 500.

DAY 1

Every memory tool remembers on day one. Every one of them. Every demo looks the same.

DAY 30

One wrong thing gets in: a bad instruction, a guess, a note nobody checked. It’s remembered as fact. Now the memory is lying to you, and you can’t tell which part.

DAY 500

Nobody trusts it, so everyone explains everything again. The memory has turned into dead weight, a hoard nobody trusts.

That’s Memory rot. It happens to every memory that keeps everything. MAGNEMO keeps only what you approve, with a receipt on every entry. On day 500, you still believe it.

How it works

Your AI plugs into MCP like USB-C plugs into your device. The one plug standard. MAGNEMO: The one plug Memory.

1Paste.One paste in Claude Code’s chat. It installs MAGNEMO and mounts it.

pip install magnemo · magnemo init · magnemo mount writes .mcp.json, the file Claude Code reads, and the first line in CLAUDE.md.

2Work.Your Agent writes drafts into Memory as you go. Nothing becomes Memory until you approve it.

Four verbs for memory, three for your yes, two for a save point. Plain files in a folder you can open.

For builders: Through the MCP server there is no write path except a draft. Every entry carries who wrote it, when, and who approved it. Trust is earned on the record and lost on a failure, never flipped by a switch.

3Come back.Close the session, open another one, anywhere. It picks up where you left it.

Your Agent wakes from its own record. Every decision still decided. Nothing forgotten.

Building fast is easy now. Building right isn’t.

Your AI can build in an afternoon what used to take a month. It can also get things wrong just as fast, and build on top of them. Speed is no longer the hard part. Knowing what’s true is. The faster you build, the more expensive one wrong memory gets. MAGNEMO is the Memory your AI keeps: only what you approve, and each one shows who approved it.

Fair questions

What’s MCP? — The USB-C of AI tools — one plug standard. Your AI plugs into helpers; MAGNEMO is the Memory plug.
I already keep notes in an Obsidian vault. — Keep your vault. MAGNEMO works beside it — plain files too. What it adds is what a folder can’t: who wrote each entry, when, and who approved it.
I already use Mem0, Zep, or another memory server. — Keep it. MAGNEMO mounts beside any memory you run and puts a witness on what gets kept.
Which tools is it tested with? — Claude Code today. Config snippets for Claude Desktop, Cursor, Windsurf, VS Code, Cline, Gemini CLI and Codex CLI ship with the source (github.com/Magnemo-AI/magnemo). Needs Python 3.11 or newer (your AI can install it).
YOUR AI WRITES DOWN WHAT IT LEARNS. YOU APPROVE WHAT IT REMEMBERS.
Sessions die. The Memory doesn’t.
The Memory that matters.

Get the notes when this changes.

Each release, and each new app MAGNEMO connects to.

RECEIPT LOGGED — you’re on the updates list.
COULDN’T LOG THE RECEIPT — try again in a moment.