From claude-twin
Capture Google Meet / Zoom captions during a live meeting (with explicit user opt-in) and produce a structured post-meeting summary — attendees, decisions, action items.
How this skill is triggered — by the user, by Claude, or both
Slash command
/claude-twin:twin-meetingsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use when a meeting is detected (event source `gmeet` or `zoom`, eventType `meeting_detected`) or when the user asks "summarise the meeting I'm in".
Use when a meeting is detected (event source gmeet or zoom, eventType meeting_detected) or when the user asks "summarise the meeting I'm in".
meeting_detected fires, ask the user for opt-in by reminding them to click Allow in the claude-twin extension popup. Captioning is off by default per the privacy policy.caption events. Display them as they arrive — do not buffer silently.meeting_ended arrives, the payload includes { captions, participants, title, startedAt }.Produce in this order:
participants.owner: action (deadline if mentioned). Pull owner from the most recent speaker before the action verb.npx claudepluginhub fadymondy/claude-twin --plugin claude-twinGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.