Keep your MEDIC fields current — without the data entry.

dealscout reads what was actually said on your calls and proposes the qualification fields in Attio. Every suggestion is cited back to the source note, and it never overwrites what a rep already entered. It just does the admin nobody wants to.

Runs on your own keys · read-only by default · for Attio

The tax reps hate paying

The MEDIC fields that make pipeline reviews useful — Metrics, Economic Buyer, Decision Criteria, Identified Pain, Champion, Paper Process, Competition — are exactly the ones that go stale. The intel exists, buried in call notes. dealscout surfaces it and proposes the fields for a human to accept. It assists; it doesn't overwrite.

How it works

1. Reads the notes. It pulls each active deal's notes — Granola call summaries and manual notes alike — straight from Attio.

2. Extracts MEDIC, with receipts. Claude pulls out the qualification signals and quotes the exact line that supports each one. No citation, no suggestion.

3. Suggests — never overwrites. Findings land in parallel scout_* fields beside the rep's own. If they've already filled a field, dealscout leaves it alone and just logs what it found.

Ask it anything in Slack

A read-only bot for call prep. @mention a deal and get its MEDIC picture, in thread — with links back to the source notes. It never writes to your CRM.

Built to be trusted against a live CRM

• It never touches a rep-owned field. Suggestions live in a parallel namespace, side by side with the real ones.
• Every value carries a citation. A verbatim quote from the note that justified it — or it isn't written at all.
• Writes are double-locked and off by default. It runs as a dry-run until you explicitly flip it on. A typo can't go live.
• Runs on your own keys. Your Attio token, your Anthropic key, your infrastructure. Nothing routes through us.
• Every action is logged. A full audit trail of what it suggested, wrote, or skipped — and why.