transcript-actions
Skill4 starsChanged 33 days agoMITView on GitHub
Extract action items, decisions, open questions, and risks from a meeting transcript or rough notes, with a short summary. Use after any client or internal call. Usage - /transcript-actions path/to/transcript.txt
In this repository11 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- change-requestSkill
- code-review-lightSkill
- deliverable-qaSkill
- exec-summarySkill
- meeting-prepSkill
- sow-reviewSkill
- stakeholder-prepSkill
- status-commsSkill
- status-reportSkill
- weekly-briefSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
If no path was given, ask for one.
- Distinctive: line 19
If your organization has connected Microsoft 365 or another meeting tool to Claude and the transcript is not on disk, you may fetch it …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Transcript Actions
Parse a meeting transcript and extract every action item, decision, open question, and risk, plus a brief summary. Saves the 20 to 30 minutes of note cleanup that follows every call.
Input
$ARGUMENTS is the path to a .txt, .md, .docx, or .vtt transcript exported from Teams,
Zoom, Otter.ai, or typed notes. Speaker labels help ("Sarah: We need to...") but are not required.
Partial or rough notes are fine.
If no path was given, ask for one.
If your organization has connected Microsoft 365 or another meeting tool to Claude and the transcript is not on disk, you may fetch it through that connector, but name the meeting and confirm with the user before reading it. Never work around a connector that is not available.
Output
Meeting Summary (3 to 5 sentences): who attended, the stated purpose, and the top outcomes or decisions.
Decisions Made:
- List each decision with context. Format: "[Decision]: [Rationale if stated]"
Action Items:
| # | Action | Owner | Due Date | Priority | Notes |
Extract all explicit and implied action items. If owner or due date is not stated, use [TBD]
and flag it. Never invent an owner or a date. Priority: High / Medium / Low based on language and
context.
Open Questions / Parking Lot:
- Items raised but not resolved during the meeting
Risks or Concerns Flagged:
- Any risks, blockers, or concerns mentioned, including informal ones
Disagreements left open:
- Any point where two people took different positions and the meeting moved on. These are the items someone will need to find later. Record positions, not impressions of people.
Be thorough. Do not summarize away specifics. The value is in the detail.
Example Output
Meeting Summary: Sprint 3 retrospective with 6 team members and 2 client stakeholders (May 6, 2025). Team reviewed velocity showing a 12% shortfall against plan. Main outcome: agreed to reduce scope on the reporting module and add one buffer week before UAT. Client stakeholders requested a revised project plan by Friday.
Action Items:
| # | Action | Owner | Due Date | Priority | Notes |
| 1 | Revise project plan with updated UAT dates | Sarah M. | Fri May 10 | High | PMO sign-off required |
| 2 | Send updated scope reduction memo to client | PM | Mon May 13 | High | Draft first, PM to review |
| 3 | Schedule UAT kick-off call with client SMEs | Coordinator | Wed May 8 | Medium | Include test script overview |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

