agentleFS
slashnote-appslashnote-app/claude-code-plugin/skills/note-meeting/SKILL.md

note-meeting

Skill1 starsChanged 9 months agoMITView on GitHub

Capture meeting notes — attendees, decisions, action items with owners

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 9 are distinctive, and 2 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create a blue note with structured format

    line 28
  2. Distinctive:

    If no clear decisions were made → omit Decisions section entirely

    line 57
  3. Distinctive:

    If no clear discussion points → omit Key Points section entirely

    line 58
  4. Distinctive:

    If user provides clear notes with bullet points, names, or decisions — organize into Decisions, Key Points, and Action Items.

    line 95
  5. Distinctive:

    If user provides a stream-of-consciousness dump:

    line 98
  6. Very specific:

    If user provides just a meeting topic (e.g., /note-meeting Sprint planning):

    line 106
  7. Distinctive:

    Create the note with the topic as title

    line 107
  8. Very specific:

    Only include Attendees: line if 1+ names detected

    line 124
  9. Distinctive:

    If more content, summarize — meeting notes should fit on one screen

    line 136
  10. Distinctive:

    Do NOT ask for more details — work with what you have

    line 217
  11. Distinctive:

    Keep it scannable — meetings generate noise, notes should be signal

    line 219

≈ “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

/note-meeting — Meeting Notes

Capture meeting notes with structured format: topic, decisions, and action items with owners.

Usage

/note-meeting <topic or raw notes>

Behavior

  1. Parse the user's input to extract:
    • Topic/title — what the meeting is about
    • Attendees — names mentioned
    • Decisions — things that were decided/agreed on
    • Key points — discussions, important info (non-decisions)
    • Action items — tasks with owners and deadlines
  2. Create a blue note with structured format
  3. Confirm: "Created meeting note: <title>"

Output Format

# <Meeting Title>
**Attendees:** Name1, Name2

## Decisions
- Decision 1
- Decision 2

## Key Points
- Point 1
- Point 2

## Action Items
- [ ] @Name: Task description (by <deadline>)
- [ ] @Name: Task description
- [ ] Task without owner

## <Date>

Section Rules

  • Decisions appears ABOVE Key Points — decisions are the highest-value output
  • Key Points — discussion items, observations, information shared (NOT decisions)
  • If no clear decisions were made → omit Decisions section entirely
  • If no clear discussion points → omit Key Points section entirely
  • At least one of Decisions or Key Points must be present

Action Item Format

Each action item should pass the "stranger test" — could someone unfamiliar with the meeting understand exactly what needs to be done?

ComponentFormatExample
Owner@Name: prefix@Maria: Update API docs
Deadline(by <date>) suffix(by Friday)
BothCombined@Maria: Update API docs (by Friday EOD)
NeitherPlain taskSet up A/B testing

Deadline Extraction

Detect deadline signals in the input and attach them to the relevant action item:

SignalOutput
"by Friday", "before Friday"(by Friday)
"by EOD", "end of day"(by EOD)
"by end of week"(by end of week)
"next week", "next Monday"(by next Monday)
"tomorrow"(by tomorrow)
"ASAP", "urgent"(ASAP)
No deadline mentionedNo suffix

Date Format

Date appears as the last section heading (not bold text):

  • Format: ## Feb 19, 2026
  • This makes it scannable in note list and separates metadata from content

Parsing Intelligence

From structured input

If user provides clear notes with bullet points, names, or decisions — organize into Decisions, Key Points, and Action Items.

From raw/unstructured input

If user provides a stream-of-consciousness dump:

  • Extract anything phrased as agreement/conclusion → Decisions ("decided to", "agreed on", "going with", "we'll use")
  • Extract anything that sounds like a task/follow-up → Action Items (with owner if name is mentioned)
  • Extract discussion topics → Key Points
  • Extract meeting topic from the most prominent subject
  • Discard filler words and organize into clean format

From topic-only input

If user provides just a meeting topic (e.g., /note-meeting Sprint planning):

  • Create the note with the topic as title
  • Add placeholder sections:
    ## Key Points
    -
    
    ## Action Items
    - [ ]
    
    ## Feb 19, 2026

Attendee Extraction

Detect names from the input:

  • Explicit: "with Alex and Maria", "attendees: John, Sarah"
  • Implicit from action items: "Maria will handle frontend" → Maria is an attendee AND action owner
  • Only include **Attendees:** line if 1+ names detected
  • Names in action items: @Maria: handle frontend changes

Formatting Rules

  • Always blue color
  • Title: concise meeting topic, no "Meeting:" prefix (the skill name implies it)
  • Decisions: bullet list — these are conclusions, not tasks
  • Key Points: bullet list — these are facts/observations, not tasks
  • Action Items: checkboxes — these are actionable follow-ups
  • Date: as ## heading at the bottom (scannable)
  • Max 5 decisions, max 7 key points, max 7 action items — keep it scannable
  • If more content, summarize — meeting notes should fit on one screen

Examples

Input: /note-meeting Sprint planning — discussed auth refactor, decided to use JWT, need to update API docs by Friday, Maria will handle frontend Result:

# Sprint Planning
**Attendees:** Maria

## Decisions
- Use JWT for authentication

## Key Points
- Discussed auth refactor approach

## Action Items
- [ ] @Maria: Handle frontend changes
- [ ] Update API docs (by Friday)

## Feb 19, 2026

Input: /note-meeting 1:1 with Alex Result:

# 1:1 with Alex
**Attendees:** Alex

## Key Points
-

## Action Items
- [ ]

## Feb 19, 2026

Input: /note-meeting We talked about the new pricing page, John suggested A/B testing, we agreed to go with tiered pricing, Sarah needs to deliver designs by Friday, also discussed the analytics bug — decided to deprioritize it Result:

# Pricing Page Discussion
**Attendees:** John, Sarah

## Decisions
- Go with tiered pricing model
- Deprioritize analytics bug

## Key Points
- Discussed new pricing page approach
- John suggested A/B testing strategy
- Analytics bug acknowledged but deprioritized

## Action Items
- [ ] @Sarah: Deliver pricing page designs (by Friday)
- [ ] Set up A/B testing for pricing page

## Feb 19, 2026

Input: /note-meeting Quick sync about deployment — we'll deploy Monday morning, Alex will run migrations first, then Sarah deploys the API, I need to update the status page ASAP Result:

# Deployment Sync
**Attendees:** Alex, Sarah

## Decisions
- Deploy Monday morning
- Migrations run first, then API deployment

## Action Items
- [ ] @Alex: Run database migrations (by Monday morning)
- [ ] @Sarah: Deploy the API (by Monday morning)
- [ ] Update status page (ASAP)

## Feb 19, 2026

Important

  • One tool call — create note immediately
  • Do NOT ask for more details — work with what you have
  • Blue color always — this is the meeting color
  • Keep it scannable — meetings generate noise, notes should be signal
  • Decisions are the #1 output — separate them from discussion
  • Action items need owners when possible — "stranger test" is the quality bar
  • Deadlines extracted from natural language, not asked for

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.