AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
← Back to list

Meeting Minutes Taker

Turns raw meeting transcripts into complete, evidence-backed structured minutes using parallel multi-pass generation and retrieval verification.

Docs & OfficeIntermediate★ 1,419⑂ 219AI score 10/10Last updated: Sep 22, 2026

What it does

  • Reads a transcript (ASR/STT output) and produces structured minutes: Executive Summary, Key Decisions, Discussion, Action Items, Parking Lot.
  • Spawns three parallel subagents, each writing a full set of minutes, then UNION-merges them so details a single pass would drop are recovered.
  • Handles anonymous speakers ("Speaker 1/2/3", "发言人1"): it first asks you to label them at the source platform (Feishu Minutes, Tencent Meeting) and re-export; text-based inference with confidence scores and user confirmation is only a fallback.
  • Retrieval Self-Test: a fresh-context agent enumerates "did X ever instruct/promise/decide Y?" claims from the transcript alone, hit-tests them against the scannable tables, runs a revocation scan, and promotes real misses with greppable verbatim quotes.
  • Also does smart date-topic-type filenames, aggressive Mermaid diagram inclusion (ER, sequence, flowchart, state), and cross-AI merging with output from Gemini/ChatGPT.

Who it's for

  • PMs, TPMs and tech leads who write minutes for recurring sprint or design-review meetings.
  • Teams that need auditable decision trails — every significant decision carries a supporting quote.
  • Anyone merging several versions of minutes or auditing existing notes for gaps.

Examples

  1. Drop in a 90-minute API design meeting transcript and ask "write the minutes" → it proposes 2026-01-25-order-api-design-review.md and produces minutes with ER and sequence diagrams.
  2. Give it your notes plus a colleague's and say "merge these" → UNION merge that removes nothing, only consolidating duplicates.
  3. Hand over existing minutes plus the original transcript and ask "what's missing?" → section-by-section comparison that restores dropped decisions, numbers and owners.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/daymade/claude-code-skills/HEAD/daymade-audio/meeting-minutes-taker/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the daymade-audio/meeting-minutes-taker folder from the GitHub repo daymade/claude-code-skills into my ~/.claude/skills/meeting-minutes-taker/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/daymade/claude-code-skills.git && mkdir -p ~/.claude/skills && cp -r claude-code-skills/daymade-audio/meeting-minutes-taker ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the repo: git clone https://github.com/daymade/claude-code-skills.git
  3. Create the skills folder if needed: mkdir -p ~/.claude/skills
  4. Copy the skill in: cp -r claude-code-skills/daymade-audio/meeting-minutes-taker ~/.claude/skills/
  5. Verify the bundled templates copied over: ls ~/.claude/skills/meeting-minutes-taker/references
  6. Restart Claude Code, then point it at a transcript file and ask "turn this into meeting minutes".
  7. (Optional) Prepare a context.md with a team directory (names, roles, speaking styles) and pass it along for far better speaker identification.