fforget

Tell Claude Code once. Cursor remembers.

One memory for your AI — decisions, preferences, project context — in a SQLite file on your machine. We couldn't read it if we tried.

For the end-to-end encrypted sync beta. No spam — launch news only.

pip install forget-ai[server] npx forget-connect # wires Claude Code, Codex, Claude Desktop # then, against any repo: $

Using an AI agent? forget's llms.txt has setup instructions for LLMs.

WORKS WITH
you — in Cursor, three weeks later

why is the payment provider Paddle and not Stripe?

⏺ forget · search_memories
Paddle was chosen as merchant of record on July 13 — KYB review passed, and it handles global tax remittance so you don't have to. [decision · 2026-07-13 · via Claude Code]

How it works

Here's what happens after npx forget-connect — with no behavior change asked of you.

  1. Install once, in sixty seconds

    One local server, one connect command. Your existing MCP setup is preserved and backed up.

  2. It remembers quietly

    Decisions surface from your conversations and your git history. The observation gate keeps facts and forgets junk — nobody writes documentation.

  3. Ask "why" from any tool

    Claude Code, Cursor, or Codex — the answer comes back with the commit or conversation that decided it.

  4. Sync arrives sealed IN DESIGN

    End-to-end encrypted, device keys, no passwords. The server — ours included — only ever holds ciphertext.

Your memory, your machine

Local by construction

Everything lives in one ~/.forget SQLite file on your machine. Search, extraction, everything — offline included.

Answers with receipts

Ask why did we choose X? — the answer links back to the commit or conversation that decided it.

Portable via MCP

The same memory in Claude Code, Cursor, and Codex. Your context follows you, not the tool.

Your model provider already reads your prompts. There is no reason your memory layer should be a second reader.

Three facts about AI memory

Memory is not a chat log. It is a compounding profile of you — and today it lives on other people's servers.

01

It is not privileged.

In February 2026, a US federal court ruled that a defendant's AI conversations were not protected — the provider's terms said data could be used for training and disclosed. What you tell an AI is, legally, what you told a company.

U.S. v. Heppner, S.D.N.Y. 2026
02

Your memory vendor will change hands.

Rewind promised local-first privacy, pivoted to the cloud, and was acquired — its users were migrated onto the acquirer's terms overnight. Every startup holding your memories is one term sheet away from being someone else's database.

Rewind → Limitless → Meta, 2025
03

The platforms will not fix this.

A model provider cannot end-to-end encrypt memory its own servers must read, and it will never port your memory to a competitor's tool. Native memory features are silos by design.

structural, not a roadmap gap

Works now. Sealed next.

SHIPPED · OPEN SOURCE

Runs on your machine today

  • One SQLite file you own — offline included
  • Mines your repo's decision history from git
  • Observation gate keeps facts, forgets junk
  • MCP: Claude Code, Cursor, Codex, Claude Desktop
IN DESIGN · E2EE SYNC

Sync that cannot betray you

  • Encrypted on your device before it touches a server
  • Embeddings encrypted too — a vector is a paraphrase
  • Scoped layers: grant a session your work memory only
  • No passwords — device keys plus a recovery code

What we won't pretend

Building AI for therapy, law, or health?

Your users' memories are your liability. A memory layer you cryptographically cannot read can't leak and can't be compelled. We're taking design partners.

Talk to us

Give your AI a memory in sixty seconds.

pip install forget-ai[server] && npx forget-connect