meeting notescrmcontactsentitiesmentionsdeal log

Meeting Scribe

Inspired by USV's Meeting Scribe agent (https://blog.usv.com/meet-the-agents), rebuilt as a generic version for any team that keeps files on people, organizations, or projects — not just a VC deal log. Reads a meeting transcript and a folder of entity files you already keep, then produces a dated meeting note, one appended mention line on each entity file the transcript actually names, and a recap email drafted for a human to review. It never guesses who a name refers to: a name either matches a file, becomes a proposed new entity for you to confirm, or is flagged as ambiguous with every candidate listed. Complements meeting-memo rather than replacing it — meeting-memo writes your circulate-ready notes, meeting-scribe keeps your people/company/project files current with a sourced mention timeline. Includes an eval contract with self-tests covering unmatched names, ambiguous names, quote-grounded mentions, and the draft-not-send boundary.

ProductivityBy uristocrat
Install

Simple install — no setup required

1

Download the skill file below

2

Add it to Claude — pick your platform:

Claude desktop app

  1. Open the Claude desktop app
  2. Go to Code, then Customize
  3. Click Create a new skill
  4. Upload the skill file you downloaded

claude.ai (web)

  1. Go to Customize, then Skills
  2. Click +, then Create a new skill
  3. Upload the skill file you downloaded
3

Say a trigger phrase to activate it

Download skill file

What this skill does

This is an overview of the skill from the site, not the skill's real instructions. The full SKILL.md Claude reads when this skill is active lives on GitHub at the pinned source below.

Inspired by USV’s Meeting Scribe agent — USV built it for VC deal logs (portfolio companies, founders, co-investors). This is a generic, independently built version, not their code: any team that keeps files on people, organizations, or projects can point this at a transcript and get the same shape of output.

Reads a meeting transcript and a folder of entity files you already keep, then produces three things: one dated meeting note, one appended mention line on each entity file the transcript actually names, and a recap email drafted for a human to review — never sent. It never guesses who “sounds like” a tracked entity. A name either matches a file, becomes a proposed new entity for you to confirm, or is flagged as ambiguous with every candidate listed. Run it after every meeting and each entity’s mention timeline compounds. Complements meeting-memo rather than replacing it — run meeting-memo for the notes your team reads, run meeting-scribe to keep your people/company/project files current.

View full SKILL.md on GitHub