agentleFS
stardustaistardustai/stardust-skills/skills/ceo-meeting-work/SKILL.md

ceo-meeting-work

Skill0 starsChanged 2 months agoMITView on GitHub

Use for meeting minutes, transcripts, silent meetings, preparation materials, summaries, decisions, and action items. Use ceo-calendar-invite for attendance decisions and ceo-work-tracking when an agreed action becomes tracked work. Load dingtalk-minutes before reading DingTalk meeting records.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 9 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use the corresponding platform Skills instead of copying their commands here.

    line 17
  2. Very specific:

    Use ceo-calendar-invite for attendance decisions and ceo-work-tracking when an agreed action becomes tracked work.

    line 27
  3. Distinctive:

    Read the meeting identity and summary first.

    line 34
  4. Distinctive:

    Read tasks when ownership, delivery, or follow-up matters.

    line 36
  5. Distinctive:

    Read transcript only when speaker attribution, disagreement, ambiguity, or an unsupported summary requires it.

    line 37
  6. Distinctive:

    Do not infer meeting content from a title, preview, link, or stale conclusion.

    line 48
  7. Distinctive:

    Do not treat a silent meeting as an ordinary notification.

    line 52
  8. Distinctive:

    Place every participant mention adjacent to that person's concrete task, question, decision, or information.

    line 59
  9. Distinctive:

    Do not assign a participant work that the meeting evidence does not support.

    line 64
  10. Very specific:

    Use canonical no_action when the meeting creates no decision, task, clarification, or useful information delivery.

    line 65
  11. Very specific:

    When an action is justified, use a canonical proposal.

    line 67
  12. Distinctive:

    If decisive evidence is absent after using the applicable operation Skills, do not guess.

    line 70

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

CEO Meeting Work

Load ceo-meeting-work for meeting preparation, records, silent meetings, summaries, decisions, participant information, and follow-up work.

Compose Operation Skills

Load the meeting operation Skill for the source platform before reading evidence. Use the corresponding platform Skills instead of copying their commands here.

SourceOperation Skills
DingTalk meeting or AI Minutesdingtalk-minutes
Lark meeting recordlark-vc and lark-minutes
DingTalk linked document or filedingtalk-doc or dingtalk-drive
Lark linked document or filelark-doc or lark-drive

Load the matching chat or document-comment Skill only when delivery requires it. Use ceo-calendar-invite for attendance decisions and ceo-work-tracking when an agreed action becomes tracked work.

Read Evidence Deliberately

  1. Identify the requested meeting outcome: preparation, summary, decision, alignment, task extraction, clarification, or information delivery.
  2. Read the meeting identity and summary first. Confirm that the record is the meeting the request refers to.
  3. Read tasks when ownership, delivery, or follow-up matters.
  4. Read transcript only when speaker attribution, disagreement, ambiguity, or an unsupported summary requires it. Read only the portions needed, but continue through all pages needed to support the judgment.
  5. Inspect linked material only when it changes or supports the requested judgment. Use the operation Skill for the material's actual platform and type.

The agent decides what evidence is needed and performs the business judgment. The service supplies references and exact read commands without interpreting meeting content. Use a supplied exact read command as given; use the loaded operation Skill to discover any additional operation required by this workflow. Do not infer meeting content from a title, preview, link, or stale conclusion.

Decide And Deliver

Do not treat a silent meeting as an ordinary notification. Silence in the source conversation does not remove a decision, task, unresolved question, or requested information found in the meeting evidence.

  • Produce only the decision, context, actions, and open questions needed by the current audience. A meeting summary is not a substitute for acting on clear tasks.
  • Place every participant mention adjacent to that person's concrete task, question, decision, or information. Never put a wall of participant mentions at the start.
  • Mention a participant once per relevant item. Do not duplicate the same mention as a heading or preamble.
  • Do not assign a participant work that the meeting evidence does not support.
  • Use canonical no_action when the meeting creates no decision, task, clarification, or useful information delivery.
  • When an action is justified, use a canonical proposal; the operation Skill owns delivery mechanics and verification.

If decisive evidence is absent after using the applicable operation Skills, do not guess. Ask for one specifically missing meeting material that the participant can supply. Do not ask for a generic meeting recap or for material the loaded operation Skill can read. If the dependency itself failed or access is unavailable, report that dependency failure and do not invent a conclusion.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.