spark-recipe-meeting-followup

name	spark-recipe-meeting-followup
description	Review a meeting transcript and draft follow-up emails with action items to attendees.
metadata	{"version":"1.0.0","requires":{"skills":["use-spark"],"accessLevel":"triage"}}

Recipe: Meeting Follow-Up

Review a meeting transcript, extract action items, and draft follow-up emails to attendees.

Prerequisite: Read the use-spark base skill for command reference and filter syntax.

Access level required: triage.

Steps

Step 1: Find the meeting transcript

spark meetings --filter "newer_than:1d"

Or search by topic:

spark meetings --filter "subject:standup"

Step 2: Review the summary

spark meeting <id>

Read the AI-generated summary for key points.

Step 3: Get full details if needed

spark meeting <id> --transcript --notes

Extract action items, decisions, and follow-up commitments from the full transcript and notes.

Step 4: Look up attendee emails

spark contacts "attendee name"

Step 5: Draft the follow-up

spark draft --to "[e-mail]" --to "[e-mail]" --subject "Follow-up: Meeting Topic - Action Items" --body "Hi team,\n\nHere are the action items from today's meeting:\n\n- [ ] Item 1 (Owner: Alice, Due: Friday)\n- [ ] Item 2 (Owner: Bob, Due: Next week)\n\nLet me know if I missed anything.\n\nBest regards"

Always confirm the draft content with the user before creating it.

Step 6: Share with team if relevant

If the action items should be visible to the team:

spark comment <related-thread-id> --body "Meeting follow-up posted - see action items in the follow-up email"

Tips

• Start with just spark meeting <id> (summary only) - the full transcript is often more detail than needed.
• The --notes flag includes any notes taken during the meeting, which often contain the most actionable items.
• Use --transcript when the summary doesn't capture something the user remembers discussing.
• Draft the follow-up as a new email (--to) rather than a reply unless there's an existing thread about the same topic.
• For recurring meetings (standups, weeklies), use spark meetings --filter "subject:standup newer_than:7d" to find the latest.