OpenLoop

AI commitment tracking for everyday conversations

Never forget what you owe — or what you're owed.

OpenLoop reads the conversations you import, finds the promises buried inside them, and keeps a two-way ledger of who owes what to whom — with the original message as proof.

Try sample conversation See the dashboard

No sign-up · Fictional sample included · Nothing is ever sent for you

openloop · analysing conversation

Group chat · Project Atlas

  • James

    I'll email you the report by Friday.

  • You

    Perfect, thanks. I'll send Sarah the slides tonight.

  • Sarah

    I might send you something next week.

    Tentative — not tracked as a promise

  • Alex

    I'll send you the API key.

3 commitments found

  • Owed to you Pending

    Email the project report

    From James Fri
  • You owe Pending

    Send Sarah the slides

    To Sarah Tonight
  • Owed to you Pending

    Send the API key

    From Alex No deadline

Todo apps remember the tasks you enter. OpenLoop remembers the promises you never entered.

“I'll send you the report tomorrow.”

Becomes a tracked promise owed to you, due tomorrow.

“Can you transfer me €20 tonight?” — “Sure.”

Becomes something you owe, with the message as proof.

“I might send you something next week.”

Stays out of your list. Maybes aren't promises.

How it works

From buried promise to closed loop in three steps.

  1. 01

    Import a conversation

    Paste a chat, drop in a screenshot or upload a .txt export from WhatsApp, Discord, Telegram or email. Tell OpenLoop which sender is you.

  2. 02

    AI finds the promises

    OpenLoop separates real commitments from maybes and small talk, then works out who owes what, to whom, and by when.

  3. 03

    Close the loop

    Track deadlines, nudge with a follow-up you've approved, and confirm when a later message shows a promise was kept.

What we're building next

Your sources. Your schedule. Less manual work.

We're moving beyond manual imports toward sources you authorise and a calendar that stays useful. The aim is less copying and pasting, with source evidence and review for uncertain details.

Planned, not yet connected: today's demo analyses conversations you paste.

Emails that become actions

Planned email connections will surface deadlines and commitments from the accounts you authorise — not just the messages you remember to copy.

University updates in one place

Planned website monitoring will help catch assignment dates, timetable changes and university events from the pages you choose.

Tickets with a place in your day

Planned ticket reading will pick out event and booking details, ready to become useful calendar entries.

A calendar backed by the source

Planned calendar entries will link back to the original email, university page or ticket, so you can check the details instead of trusting a guess.

Close your open loops.

Import one conversation and see every promise inside it in under a minute.

Get started