session-debrief

Converts Granola workshop transcripts into session summaries and appended Q&A entries.

11|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/qa-aman/next-leap-claude-code --skill session-debrief-qa-aman
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: session-debrief
Source: https://github.com/qa-aman/next-leap-claude-code/tree/main/.claude/skills/session-debrief
Command: npx skills add https://github.com/qa-aman/next-leap-claude-code --skill session-debrief-qa-aman

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After each live workshop session, manually turning a long Granola transcript into a structured session summary and updating a running Q&A document is slow and error-prone. This Skill automates that post-session workflow so both artifacts stay consistent in format, numbering, and citation quality. ## Core Features & Use Cases - Transcript Processing: Fetches a Granola meeting transcript via the Granola MCP tool and reads it in chunks to handle very long content. - Session Summary Generation: Writes a structured summary file (topics covered, audience profile, key quotes, action items, session rating) under .local/session-transcripts/ mirroring the most recent session's format. - Q&A Continuation: Appends 10-15 numbered Q&A entries to 02-presentation/q&a.md, continuing existing numbering and validating every cited URL. - Use Case: After teaching a bootcamp session, paste the Granola link and say "session completed" to get a summary file and updated q&a.md with verified sources. ## Quick Start Say "session completed, here is the Granola link" followed by the Granola URL to generate the session summary and append the new Q&A entries.

Frequently Asked Questions about session-debrief

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

FAQPage Schema
How do I process a Granola transcript after a workshop session?▼

Paste the Granola link with a phrase like "session completed" or "debrief this session". The skill extracts the meeting UUID from the URL, fetches the transcript via the Granola MCP tool, and generates the summary and Q&A updates.

How to update a running Q&A document with new session questions?▼

The skill reads the existing q&a.md to find the last question number and heading format, then appends a new dated section with 10-15 numbered questions, short answers, and verified source URLs.

Can I use this with transcripts from tools other than Granola?▼

No, the workflow depends on the Granola MCP get_meeting_transcript tool and a Granola URL containing the meeting UUID. Other transcript sources would require a different fetch mechanism.

Why does the skill read transcripts with Python instead of the Read tool?▼

Granola transcripts exceed 100k characters with very long lines, which the Read tool handles poorly. The skill uses Python character-range slicing to process the transcript in manageable chunks.

What are the limitations of the generated Q&A citations?▼

Sources are restricted to official Anthropic and Claude documentation domains, and every URL is validated with curl before writing. If no first-party page exists for a topic, the citation is omitted rather than substituted with a blog.