Meeting Insights Analyzer

Analyzes meeting transcripts to surface behavioral patterns, communication anti-patterns, and actionable coaching feedback. It produces a structured report with findings on speaking dynamics, directness, filler words, question quality, facilitation, and sentiment.

✨ What it does

  • Ingests and normalizes transcripts from various formats (VTT, SRT, plain text, DOCX, JSON).
  • Calculates per-speaker speaking dynamics: word count, turn count, average turn length, longest monologue, interruptions.
  • Detects hedging language and conflict avoidance patterns with severity tags and rewrite suggestions.
  • Counts filler words and computes rate per 100 words, flagging only if above threshold.
  • Classifies questions (closed, leading, open, clarifying, building) and assesses listening indicators.
  • Evaluates facilitation quality (agenda adherence, time management, inclusion, decision clarity, action items) when user

🎯 When to use it

  • User uploads or points to meeting transcripts (.txt, .md, .vtt, .srt, .docx) and wants feedback on communication habits.
  • User asks about how they run meetings, speaking ratio, filler words, or conflict avoidance.
  • User wants to compare communication across time periods or mentions tools like Granola, Otter, Fireflies, or Zoom transcripts.
  • User says 'look at my meetings' or 'how do I come across in meetings'.

🚀 How to use

Trigger by uploading or referencing meeting transcript files, or by asking for feedback on meeting communication. The skill expects transcript files in a directory or provided path. It will ask for user identity if ambiguous. Example prompts:

Analyze my meeting transcripts in ./meetings and give me feedback.
How do I come across in meetings? Here are my Zoom transcripts.

📄 Output: A markdown report with top findings, detailed analysis sections, strengths, growth opportunities, and optional comparison to previous period.

📦 Add this skill to Claude Code

# 1. Get the skills repo
git clone --depth 1 https://github.com/alirezarezvani/claude-skills /tmp/claude-skills

# 2. Copy this skill into your project (or ~/.claude/skills for all projects)
mkdir -p .claude/skills
cp -r /tmp/claude-skills/project-management/skills/meeting-analyzer .claude/skills/meeting-analyzer

Skill source: project-management/skills/meeting-analyzer/SKILL.md

⚠️ Good to know

Requires speaker labels for per-speaker metrics; filler word and hedging dictionaries are English-centric; single-meeting analysis is less reliable for trends.

❓ FAQ

What if my transcript has no speaker labels?

The skill will warn you and run text-level analysis only (filler words, question types) but skip per-speaker metrics. It suggests re-exporting with speaker diarization enabled.

Can I analyze a single meeting?

Yes, but the skill will caveat that patterns from short meetings may not be representative and will skip trend/comparison language.

Does the skill support non-English transcripts?

It notes that filler word and hedging dictionaries are English-centric, so for other languages it focuses on structural analysis like speaking ratios and turn-taking.

🤖 Overview, features, install steps and FAQ were generated from the project's SKILL.md on Sep 4, 2026. Always check the original source before running commands.