agentleFS
jaredhhawkjaredhhawk/claude-skills/skills/meetings/SKILL.md

meetings

Skill0 starsChanged 8 months agoMITView on GitHub

Review synced meeting notes. Finds unprocessed notes, suggests routing, and reports sync status.

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. 0 are common, 11 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 11

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read the last 10 lines of 99_System/scripts/sync_granola.log and report:

    line 14
  2. Distinctive:

    When the last sync ran

    line 15
  3. Distinctive:

    Read the note content and suggest:

    line 28
  4. Very specific:

    Update Domain in frontmatter

    line 39
  5. Very specific:

    Remove processed: false if present

    line 40
  6. Very specific:

    Add or update Related: line after frontmatter

    line 41
  7. Distinctive:

    If subfolder was approved and differs from current location, move the file

    line 42
  8. Distinctive:

    Add entity wikilinks in the content body where appropriate

    line 43
  9. Distinctive:

    After processing all notes, show:

    line 47
  10. Very specific:

    Use these rules when suggesting Domain and subfolder (loaded from 99_System/scripts/granola_routing.json):

    line 54
  11. Distinctive:

    When processing note content, convert recognized mentions to wikilinks:

    line 68
  12. Distinctive:

    Only link first occurrence in each section.

    line 74
  13. Distinctive:

    Start with sync status, then list unprocessed notes

    line 78
  14. Distinctive:

    If no unprocessed notes, just report status and exit

    line 79
  15. Distinctive:

    If user says "approve all", batch-apply all suggestions

    line 81
  16. Distinctive:

    If user says "skip", move to next note

    line 82

“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “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

/meetings - Meeting Notes Review

You are helping the user review and process meeting notes synced from Granola.

Behavior

1. Report sync status

Read the last 10 lines of 99_System/scripts/sync_granola.log and report:

  • When the last sync ran
  • How many notes were created in the last run

2. Find unprocessed notes

Scan 99_System/Meeting Notes/ (including subdirectories) for notes that have granola_id in frontmatter AND either:

  • processed: false in frontmatter, OR
  • Missing Domain field in frontmatter

List them with title, date, and current location.

3. For each unprocessed note

Read the note content and suggest:

  • Domain: Which Domain MOC to link (e.g., [[D - Career & Job Search]])
  • Related links: Relevant wikilinks based on content (people, projects, etc.)
  • Subfolder: Where it should live under 99_System/Meeting Notes/
  • Entity links: Convert plain text mentions to wikilinks in the content body

Present suggestions to the user and ask to approve / edit / skip each one.

4. Apply approved changes

For each approved note:

  1. Update Domain in frontmatter
  2. Remove processed: false if present
  3. Add or update Related: line after frontmatter
  4. If subfolder was approved and differs from current location, move the file
  5. Add entity wikilinks in the content body where appropriate

5. Summary

After processing all notes, show:

  • How many notes were processed
  • How many were skipped
  • Any that need manual attention

Routing Reference

Use these rules when suggesting Domain and subfolder (loaded from 99_System/scripts/granola_routing.json):

Keywords / PeopleDomainSubfolder
Mark R., JaredMark[[D - Career & Job Search]]Mark_R.
Nate B. Jones, NateBJ[[D - Tech & AI]]NateBJones
Legata, Alesia, Bryan, Tim[[D - Career & Job Search]]Legata
Theron, Mitch, branding, website, HPM[[D - Property Management]]HPM
Omaya, admissions[[D - Family]](root)
SaaS, playbook[[D - Side Hustles]](root)
Equip, outreach[[D - Career & Job Search]]Outreach_Calls

Entity Linking

When processing note content, convert recognized mentions to wikilinks:

  • "Legata" / "Alesia" / "Bryan" → [[Legata Index|Legata]] etc.
  • "Nate" / "Nate B" → [[People Index|Nate]]
  • "Theron" / "Mitch" → [[People Index|Theron]] etc.
  • "Mark R" / "Mark Reveley" → [[People Index|Mark R.]]

Only link first occurrence in each section. Don't link inside transcript sections.

Conversation Style

  • Start with sync status, then list unprocessed notes
  • If no unprocessed notes, just report status and exit
  • Be concise - show suggestions as a table or short list
  • If user says "approve all", batch-apply all suggestions
  • If user says "skip", move to next note

Example

Claude: Last sync: 2 min ago (3 notes created).

Found 2 unprocessed notes:

1. **Prop Tech** (2025-12-09) - currently in Meeting Notes root
   Suggest: Domain [[D - Tech & AI]], subfolder: none, Related: [[Meeting Index]] | [[People Index]]

2. **continuing product management conversation** (2026-01-29) - currently in Legata/
   Suggest: Domain [[D - Career & Job Search]], Related: add [[Legata Index]]

Approve all, or review individually?

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.