Minutes Mirror

Most people have no idea what they actually sound like in meetings. This skill runs the numbers on your own behavior: talk time ratio, filler words, hedging language, monologue length, and how long you go without interrupting. It compares each meeting against your 30-day baseline so you see what was unusual about you, not just what happened. If you've tagged meetings with /minutes-tag, it will show you what you do differently in meetings you win versus ones you lose. The output is one concrete thing to try next time, not a personality audit. It uses a Python script for the counting because LLMs are terrible at exact token math. Trigger it with "how did I do" or "coach me" or just "mirror."

GitHub stars: 1.3K
last updated: Jun 11, 2026

View on GitHub

Install

$ npx -y skills add silverstein/minutes --skill minutes-mirror --agent claude-code

Installs into .claude/skills of the current project.

Files

/minutes-mirror

Self-coaching analysis based on your own meeting transcripts. Two modes:

Single-meeting mode — review a specific meeting and surface what you did, what was unusual for you, and one concrete thing to try next time.
Pattern mode — surface trends across the last 30 days, including (if meetings are tagged) what behaviors correlate with winning vs losing.

The point is not to roast you. The point is to give you a kind, evidence-based mirror to behaviors that are usually invisible to you because you're inside them.

How it works

Phase 0: Identify "you"

Mirror needs to know which speaker label in the transcript is the user. Real transcripts use one of two formats:

Enrolled users: [Mat 0:00] Hey there. — first-name labels from voice enrollment
Non-enrolled users: [SPEAKER_0 0:00] Hey there. — generic labels from diarization

Either way, mirror needs to know which label maps to the user. Check sources in order:

1. Enrolled voice profile: