Transcript Critic

Analyze meeting transcripts to identify decision-making flaws and generate structured critique reports.

Updated May 11, 2026
One-click install
npx skills add https://github.com/rafalwizen/plugin-architecture-test --skill transcript-critic-rafalwizen
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Transcript Critic
Source: https://github.com/rafalwizen/plugin-architecture-test/tree/main/week8/1/transcript-critic
Command: npx skills add https://github.com/rafalwizen/plugin-architecture-test --skill transcript-critic-rafalwizen

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Meeting transcripts often hide critical decision-making flaws like false consensus, marginalized voices, and unexamined opinions treated as facts, which lead to poorly grounded decisions if left unchecked.

Core Features & Use Cases

  • 7-Point Analysis Framework: Evaluates transcripts for fact vs opinion classification, genuine consensus, interrupted topics, hidden dependencies, scope drift, severity mismatches, and unfair authority dynamics.
  • Structured Critique Reports: Generates findings with severity ratings, verbatim evidence quotes, and targeted diagnostic questions for follow-up meetings.
  • Use Case: After a cross-team product roadmap meeting, use this skill to audit if all stakeholders genuinely agreed to the prioritization, or if junior engineers' concerns about technical debt were dismissed without review.

Quick Start

Use the transcript-critic skill to analyze the attached meeting transcript from the Q4 roadmap planning session and produce a structured critique report with findings and questions for the next team sync.

Frequently Asked Questions about Transcript Critic

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit meeting transcripts for false consensus and hidden decision flaws?

To audit meeting transcripts for false consensus, you analyze the text to identify marginalized voices and unverified opinion-to-fact escalation. This process generates structured critique reports with severity ratings and verbatim evidence quotes to validate post-meeting decisions.

What is opinion-to-fact escalation in meeting reviews?

Opinion-to-fact escalation in meeting reviews happens when unexamined subjective statements are accepted as objective facts during decision-making. Transcript analysis isolates these instances, providing verifiable diagnostic questions to prevent poorly grounded decisions in subsequent team syncs.

How do I check stakeholder meeting notes for scope drift and hidden dependencies?

You check stakeholder meeting notes for scope drift and hidden dependencies by applying a 7-point analysis framework to the transcript. This evaluates interrupted topics and unfair authority dynamics, producing a structured critique report with targeted diagnostic questions for follow-up meetings.

Can I use transcript critique to prepare diagnostic questions for a follow-up meeting?

Yes, you can use transcript critique to prepare diagnostic questions for a follow-up meeting by analyzing the previous session's transcript. The critique identifies severity mismatches and genuine consensus levels, outputting targeted questions to improve subsequent decision-making quality.

Does meeting transcript analysis work for reviewing third-party meeting notes across all team contexts?

Meeting transcript analysis works for reviewing third-party meeting notes across all team and stakeholder contexts. It evaluates fact vs opinion classification and unfair authority dynamics, generating structured critique reports with verbatim evidence quotes regardless of the meeting source.

What is the best way to validate cross-team product roadmap prioritization decisions?

The best way to validate cross-team product roadmap prioritization decisions is to audit the meeting transcript for false consensus and dismissed technical debt concerns. This analyzes unfair authority dynamics, producing diagnostic questions to ensure junior engineers' inputs are genuinely reviewed.