From zoom-suite
Live Zoom meeting debrief — automatically pulls any meeting from Zoom by name, topic, or recency, retrieves the AI summary, transcript, and next steps, then delivers a structured debrief: decisions made, action items with owners and deadlines, follow-ups, and key insights. Saves to Cloud Brain and creates a Zoom Doc with the structured output. No pasting required. Triggers on: "debrief my last meeting", "process the call with [name]", "what happened in my meeting about [topic]", "pull up the recording from [date]", "meeting debrief", "summarize my Zoom call", "extract action items from my meeting", "post-meeting summary", "what did we decide in the [topic] meeting".
How this skill is triggered — by the user, by Claude, or both
Slash command
/zoom-suite:comm-zoom-debriefThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
> **Requires:** Zoom MCP connector. All data is pulled live from Zoom — no transcript pasting needed.
Requires: Zoom MCP connector. All data is pulled live from Zoom — no transcript pasting needed.
This skill finds any meeting in the user's Zoom account, pulls the full AI summary, transcript, and auto-generated next steps, then structures everything into a professional debrief: decisions made, action items with owners and deadlines, follow-ups needed, and key insights. The structured debrief is saved to Cloud Brain and optionally pushed back into Zoom as a Zoom Doc the whole team can reference.
This replaces the manual workflow of copying transcripts out of Zoom and pasting them into a separate tool.
Before locating the meeting, load any saved communication preferences from Cloud Brain.
Search Cloud Brain using mcp__cloud-brain__search_notes with query: "zoom debrief preferences".
If preferences ARE found: Confirm in one line — "Using your saved preferences — [name], output style: [style]."
If NO preferences found: Ask these two quick questions before proceeding:
Save to Cloud Brain:
mcp__cloud-brain__write_notebrain/preferences/zoom-debrief-preferences.mdDetermine which meeting to process from the user's request.
Ask for user's current timezone before searching — this is required by the Zoom search API. If the user has preferences saved that include timezone, use that. Otherwise ask: "What timezone are you in? (e.g., America/Phoenix, America/New_York)"
Parse the user's request for meeting identifiers:
Search for the meeting using mcp__zoom__search_meetings:
q to the topic/keywords or person's namefrom/to based on the date context (if "last week," set the range accordingly)from to 7 days agoIf multiple meetings match: Display a brief list and ask the user to confirm which one:
I found [N] meetings that might match:
1. [Topic] — [Date] — [Duration]
2. [Topic] — [Date] — [Duration]
3. [Topic] — [Date] — [Duration]
Which meeting would you like to debrief?
If no meeting matches: Ask the user to refine — "I couldn't find a match. Can you give me more detail about the meeting topic or date?"
Using the meeting_uuid from the search result, call mcp__zoom__get_meeting_assets.
This returns:
meeting_summary — Zoom AI Companion's auto-generated summary (quick recap, full text, next steps)my_notes — Host/participant notes taken during the meetingrecording — Cloud recording URL and processing statusparticipants — Attendee listagenda_doc — Agenda document if one was attachedIf recording is still processing (processing: true), note: "The recording is still being processed by Zoom. The AI summary and notes are available — I'll use those. If you want the full transcript, try again in a few minutes."
If recording is available, call mcp__zoom__get_recording_resource with types: "transcript,summary,nextStep" to get:
Use both the AI summary AND the transcript. The summary is fast; the transcript catches specifics that the summary may gloss over.
If transcript is unavailable (recording not enabled, still processing, or no recording for this meeting): Proceed with AI summary and notes only. Flag: "No transcript was available — analysis is based on Zoom's AI summary and meeting notes."
Analyze all available data (AI summary, Zoom next steps, transcript, notes) and extract:
For each action item, capture:
| Field | What to Extract |
|---|---|
| Action | What needs to be done (specific, clear) |
| Owner | Who is responsible (name or role) |
| Deadline | Extract if mentioned; apply default if user has one saved; otherwise "TBD" |
| Priority | High / Medium / Low based on urgency signals in conversation |
| Source | Brief quote or context from transcript where this was stated |
Highlight action items assigned to the user (by name match against saved preferences). These appear first in the output under "Your Action Items."
Content to EXCLUDE:
Format the structured debrief:
# Meeting Debrief: [Meeting Topic]
**Date:** YYYY-MM-DD
**Attendees:** [Name 1], [Name 2], [Name 3]
**Duration:** [X minutes]
**Source:** Zoom recording — [meeting_uuid]
---
## Summary
[2–4 sentence executive summary of what the meeting was about and what was accomplished]
---
## Your Action Items
| # | Action | Deadline | Priority |
|---|--------|----------|----------|
| 1 | [Task — for the user specifically] | [Date/TBD] | High |
## All Action Items
| # | Action | Owner | Deadline | Priority |
|---|--------|-------|----------|----------|
| 1 | [Task] | [Name] | [Date/TBD] | [Level] |
| 2 | [Task] | [Name] | [Date/TBD] | [Level] |
---
## Decisions Made
| # | Decision | Made By |
|---|----------|---------|
| 1 | [What was decided] | [Person] |
---
## Follow-Ups
- [ ] [Item] — Owner: [Name] — By: [Date/Next meeting]
- [ ] [Question to resolve] — Owner: [Name]
---
## Key Insights
- [Strategic observation or important context]
- [Notable commitment, concern, or relationship insight]
---
*Processed by COMM-zoom-debrief. Source: Zoom meeting [meeting_uuid] on [date].*
Save the structured debrief to Cloud Brain:
mcp__cloud-brain__write_notebrain/meetings/[slug-from-topic]-[YYYY-MM-DD].mdAlso check if there is an active project in the brain related to this meeting's topic:
mcp__cloud-brain__search_notes with the meeting topic keywords## Meeting Update [YYYY-MM-DD]
- [1–2 sentence summary of relevant outcomes]
- Action items: [N items assigned]
After saving to Brain, ask: "Want me to also save this debrief as a Zoom Doc so your team can see it?"
If yes, call mcp__zoom__create_new_file_with_markdown:
file_name: "Meeting Debrief — [Topic] — [YYYY-MM-DD]"content: the full structured debrief in markdownProvide the returned file_link to the user.
Report to the user:
Debrief complete: [Meeting Topic] — [Date]
Attendees: [N] people
Extracted:
✅ [N] decisions
✅ [N] action items ([X] assigned to you)
✅ [N] follow-ups
✅ [N] key insights
Saved to: brain/meetings/[slug]-[date].md
[Zoom Doc link if created]
Your next actions:
1. [User's #1 action item] — [Deadline]
2. [User's #2 action item] — [Deadline]
npx claudepluginhub owenmecham/mbg --plugin zoom-suiteGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.