Scan recent calendar meetings, find new attendees, look up their LinkedIn profiles, and present connection links. Use this skill whenever the user asks to find people they've met with recently who they should connect with on LinkedIn, or asks to check calendar meetings for networking follow-ups, or mentions connecting with meeting attendees on LinkedIn, or asks "who should I connect with" after meetings. Also triggers for variations like "run my LinkedIn connect task", "check my calendar for new contacts", or "networking follow-up from meetings".
Scheduled Gmail contact extraction — scans recent emails, pulls out names, emails, phone numbers, organizations, and titles from headers, forwarded messages, signatures, and body text, then saves everything to a deduplicated spreadsheet. Use this skill whenever someone wants to automatically build or maintain a contact list from their email, extract contact info from Gmail, or set up a recurring task that harvests new contacts into a spreadsheet.
Triggered when the user spots an error in Claude's output and wants a thorough audit to find all other errors. Use this skill whenever: a user points out a mistake (wrong timezone, misspelled name, incorrect number, broken link, wrong variable, bad reference, etc.) and wants confidence that no other errors remain; a user says "brown M&M", "canary", "what else is wrong", "audit this", "check everything", "I found a mistake — what else did you get wrong", "this timezone is wrong, what other details are off"; a user expresses distrust of an output's accuracy after finding one flaw; a user asks you to do a careful review or double-check of something you produced. This skill applies to ANY output type: emails, code, documents, research, slide decks, data, plans, or anything else. Even if the user only mentions one small error, treat it as a signal that a full audit is warranted.
Use when you've just completed work and the user wants an honest audit of it — confidence levels, shortcuts taken, unverified assumptions, and known weaknesses. Triggers on: "confess", "be honest about this", "what are you unsure about", "what did you cut corners on", "what's the weakest part", "audit your own work", "what assumptions did you make", or after any significant output where the user wants unvarnished reflection rather than reassurance.
Use this skill whenever someone shares notes, a transcript, or a summary from an event, webinar, panel, training, workshop, study hall, or open mic and wants it turned into a shareable recap. Triggers on: "write a recap", "summarize this event", "turn this into a blog post", "I need a writeup of this webinar", "recap this for the newsletter", or when event/webinar notes or transcripts are shared with intent to publish or distribute. Also triggers when someone asks for social media posts about an event that happened.
Use this skill whenever someone shares meeting notes, a transcript, or a recording summary and wants it processed into structured output. Triggers on: pasted transcripts (Zoom, Circleback, Otter, Google Meet, Teams), raw meeting notes, or requests like "what were the action items", "summarize this meeting", "pull out the decisions", "what did we agree on", or "process these notes". Also triggers when a .vtt, .txt, or .srt transcript file is shared.
Use this skill whenever a URL is shared and needs to be formatted as a resource entry for a curated resource guide. Triggers when someone pastes a link to an article, report, tool, training, newsletter, blog, or any web resource and wants it formatted consistently. Also triggers on phrases like "format this link", "add this to the guide", or "here's a new resource". Activate proactively when a bare URL is shared without explanation.