What it does

Meeting feedback usually rests on memory, so this skill starts from the transcript instead. It ingests VTT, SRT, plain text, Markdown, DOCX and JSON exports, inventories dates, durations and speakers, normalises everything into one structure, and then runs six independent modules: speaking dynamics, conflict and directness, filler words, question quality and listening, facilitation and decision making, and sentiment. Where the format allows it reports real numbers, and the closing report lists findings, strengths, growth areas and comparisons with earlier meetings.

Works with

Documentation-only skill; it works with any prompt-capable agent and needs no runtime environment.

Source and safety

Listing-time review: no scripts, documentation only. (This review was conducted when the skill was listed, not as ongoing monitoring. Sources are beyond our control — evaluate skills before using them.)