Skip to content
Clone-Me

Features

Tools that connect your AI agent, and a console to review and manage memories. Upcoming features are marked “Planned”.

AI agent connection

Connect an AI agent such as Claude Code with a single connection token. The agent uses three MCP tools to save memories, look them up and ask for judgments.

  • rememberSave memories
  • recallLook up memories
  • ask_meAsk for a judgment
  • remember: sends the moments you choose, correct the AI or give a reason as cases, exactly as they happened, plus important information about you. Your AI agent doesn't write principles; the server reads cases across conversations and writes them. Nothing is a memory until you approve it.
  • recall: returns confirmed memories with their source, validity period, conditions and exceptions.
  • ask_me: returns a judgment and its evidence for a question or plan. When there's no basis, it holds off and returns a question to ask you.
  • Connections use HTTP MCP. Any client that supports an Authorization header can connect the same way.

Inbox

See the proposals the server raised from the cases your AI agent left, and approve, edit or reject them in one place. Nothing becomes memory unless you choose it.

  • Each proposal shows its content, its source, why it came up and the cases behind it (your own words), and the suggested scope.
  • The default is “not selected”. Nothing is approved by silence or a pre-checked box.
  • Select several proposals to approve or reject them at once.
  • Owner requests from dreaming are answered on the Owner requests tab, available once dreaming opens.

Memory management

Browse confirmed memories by type, and edit, archive or withdraw them at any time.

  • Edits are saved as new versions, so the history stays.
  • A withdrawn memory leaves your judgments immediately.
  • Add or pin memories yourself, and compare versions.

Dreaming

Planned

Periodically reviews stored information, decision cases and principles to keep your memory current. It does not train the AI model itself.

  • Only changes that add evidence under the same conditions are applied automatically.
  • Everything else comes to your inbox as an owner request for you to decide.

Connections and usage

Create a separate connection for each computer and AI agent, and issue, reissue or revoke connection tokens.

  • The token is shown once, right after it's issued, and never stored in plain text.
  • Disconnect a single connection, or all of them at once if you suspect a leak.
  • Usage and credit views are in progress.

Stop re-explaining. Work with an AI that knows your criteria.

Sign-up is invite-only for now. Leave your email and we'll send an invite as soon as a spot opens.

Join the waitlistAlready invited? Log in