capture-transcript

Convert pasted meeting transcripts into standup summaries or memory records.

1|Updated May 15, 2026
One-click install
npx skills add https://github.com/mslshao/claude-harness --skill capture-transcript
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: capture-transcript
Source: https://github.com/mslshao/claude-harness/tree/main/dotclaude/skills/capture-transcript
Command: npx skills add https://github.com/mslshao/claude-harness --skill capture-transcript

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents important details from getting lost when meeting, standup, sync, or 1:1 transcripts are pasted into chat. It helps convert messy spoken transcripts into either a fast action-oriented breakdown or a durable memory record with the right indexing and privacy handling.

Core Features & Use Cases

  • Transcript classification and routing: Detects whether a transcript is a standup, sync, org meeting, or 1:1, then sends it through the correct capture workflow.
  • Action extraction and durable memory capture: Produces a concise standup action summary in chat or creates a structured memory file, recall bead, log entry, and index row for meetings worth preserving.
  • Privacy and identity safeguards: Resolves names against org context, avoids fabrication, keeps 1:1 sensitive content local when required, and enforces formatting constraints for memory hygiene.
  • Use case: After a product-engineering sync or a candid 1:1, use this Skill to preserve decisions, signals, and action items in a searchable memory system without manually applying filing conventions.

Quick Start

Ask the AI to capture this sync or 1:1 after pasting the transcript, and it will summarize it and store it using the correct memory workflow.

Frequently Asked Questions about capture-transcript

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

FAQPage Schema
How do I extract action items from meeting transcripts automatically?

To extract action items from meeting transcripts, paste the transcript text into chat and ask the AI to capture it. The system classifies the meeting type and generates a concise action-focused breakdown for standups or a durable memory record for syncs.

Can I save 1on1 transcripts as memory without exposing sensitive content?

Yes, you can save 1on1 transcripts as memory safely. The capture workflow applies sensitivity gating for private 1:1 content, resolving names against org context and keeping sensitive details local when required to maintain privacy.

What is the best way to organize standup notes for future recall?

The best way to organize standup notes for recall is converting transcripts into structured memory files. This process creates recall beads, appends log entries, and updates indexes, allowing you to search past decisions and signals without manual filing.

Does meeting memory capture work for different types of syncs and org meetings?

Meeting memory capture works effectively for different syncs and org meetings. The system detects the transcript type through classification routing, then sends the content through the correct workflow to apply the appropriate summarization, indexing, and privacy-aware storage.

How do I prevent important details from getting lost in messy spoken transcripts?

To prevent details from getting lost in messy spoken transcripts, use an automated capture workflow to parse the text. It converts unstructured meeting dialogue into structured memory records, extracting decisions and action items while enforcing formatting constraints for memory hygiene.