pick-meetings-to-prep
Skill7 starsChanged 49 days agoMITView on GitHub
Asks which of today's meetings you want to prep for (you name them - no calendar read by default), then runs the prep-meeting skill once per meeting you name. Standalone, or run as a step from start-the-day. Trigger: /pick-meetings-to-prep, or asking which meetings to prep.
- Installs packages
In this repository12 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- end-taskSkill
- end-the-daySkill
- link-taskSkill
- log-completed-tasksSkill
- prep-meetingSkill
- search-mailSkill
- setupSkill
- start-the-daySkill
- suggest-brag-itemsSkill
- surface-stale-tasksSkill
- update-tasksSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
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 43
If you'd rather Claude read your calendar and offer a multiselect, you can swap step 1 for a calendar read.
- Distinctive: line 68
Never invent meetings or tasks.
- Distinctive: line 69
Never create or modify calendar events.
≈ “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
Pick Meetings to Prep
Purpose
Get the list of meetings you want to prep for today and hand each one off to
prep-meeting. This skill only collects the list - the actual prep work
belongs to prep-meeting.
By default you just name the meetings yourself. The author originally had Claude read the calendar automatically (see "Optional: an automated calendar read" below), then dropped it: it broke when moving from the terminal to the Claude desktop app, and chasing a fix was more effort than saying what's on. Naming two or three meetings takes seconds, and you're the best judge of which ones need prep anyway.
Trigger
/pick-meetings-to-prep, or asking which meetings to prep. Also invoked as a
step by /start-the-day.
Process
- Ask which meetings you want to prep for today. A plain question in chat - reply with names or topics (and times if you like). Don't guess or suggest meetings.
- If the answer is none, say so and stop.
- For each meeting named, run the
prep-meetingskill. That skill handles finding/creating the meeting note and linking it into today's Daily note - this step is just the loop that invokes it once per named meeting.
Optional: an automated calendar read
If you'd rather Claude read your calendar and offer a multiselect, you can swap step 1 for a calendar read. What the author learned trying this on macOS:
icalBuddy(brew install ical-buddy) reads events from Calendar.app, including a Google Calendar synced into it:icalBuddy -ic "<calendar name>" -nc -iep "title,datetime" -b "- " -df "" -tf "%H:%M" eventsTodayIt expands recurring events to today's actual occurrence.- Plain AppleScript against Calendar.app is unreliable here - it returned a recurring event's original start date rather than today's occurrence, so date filtering silently returned the wrong events.
- macOS grants calendar access per app.
icalBuddyworked from Terminal, but from the Claude desktop app it saw no calendars at all ("No calendars"), because the desktop app doesn't request calendar permission and so never appears in System Settings to be granted. If you use the desktop app, expect this route not to work. - Fetching a Google Calendar secret ICS feed via a web-fetch tool misdated recurring events, because the fetch summarises the feed rather than parsing it. Avoid it for anything date-sensitive.
Whatever method you use, if the read fails, say so rather than silently falling back to asking - make the fallback an explicit choice.
Guardrails
- Never invent meetings or tasks.
- Never create or modify calendar events.
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.

