Analyze meeting transcripts for talk-time, filler, and hedging metrics.
npx skills add https://github.com/silverstein/minutes --skill minutes-mirror
Please help me install this Agent Skill. Skill: minutes-mirror Source: https://github.com/silverstein/minutes/tree/main/tooling/skills/sources/minutes-mirror Command: npx skills add https://github.com/silverstein/minutes --skill minutes-mirror
Minutes Mirror gives you a clear, evidence-based view of your behavior in meetings so you can identify patterns that are invisible while you're inside the conversation. It removes guesswork about talk-time, filler words, hedging, and long monologues so you can improve specific behaviors rather than generic advice.
Ask the assistant to review your last meeting and give a concise, evidence-based one thing to try.
High-intent search queries and answers about installing and using this skill.
Meeting transcript analysis measures your talk-time and filler words by running deterministic Python scripts against diarized transcripts to calculate talk ratios, filler counts, and longest monologues. The script processes the transcript text, identifies your speaker label, and surfaces concrete metrics.
Meeting pattern analysis aggregates metrics across a specified time window like 30 days to surface behavioral trends, time-of-day fatigue, and topic correlations. When meetings are tagged with outcomes like won or lost, it correlates your talk-time and hedging rates with those results.
Diarized transcripts support self-coaching without an enrolled voice profile by locating your speaker label via cached self names or a one-time prompt. If diarization is missing entirely, the tool falls back with clear error messages.
Reducing hedging and filler words requires evidence-based self-coaching that measures your exact filler and hedging rates from transcripts. By reviewing single meetings, you get 2-3 concrete moments to re-read and one testable suggestion to try in your next conversation.
Meeting behavior analysis relies on bundled Python scripts to deterministically process diarized transcripts and measure talk-time, question frequency, and monologue length. The scripts handle speaker identification, caching, and threshold surfacing automatically.
Transcript analysis fails when diarization is missing because the Python scripts require distinct speaker labels to isolate your behavior and measure talk-time. Without diarization, speaker identification cannot proceed, and the tool returns a clear error message.
DanialHaseebTranscribe audio files into text or speaker-labeled JSON segments.
run6270Analyze meeting transcripts to detect communication patterns and generate feedback.
silversteinTag meeting records with outcome labels and edit YAML frontmatter atomically.
ricableAnalyze meeting transcripts to detect speaking ratios, filler words, interruptions, and cite evidence.
