Easy to start with

Instantly resolves CDN URLs and downloads brand icons for AI models and providers like Claude, GPT and Gemini.

Instantly sends a single plain-text message to a WeCom (Enterprise WeChat) group bot via a stored webhook.

Exports your Douban book/movie/music/game history to Excel-ready CSV files via the Frodo API, with optional RSS incremental sync.

Reverses repomix packing by extracting XML/Markdown/JSON bundles back into their original files and folders.

Configures a WeCom (Enterprise WeChat) group-bot webhook and sends clear, well-structured status, backup, and alert notifications.

Pull subtitles from any YouTube video as timed JSON, plain text, SRT or VTT in a single call.

Pull timestamped closed captions from any public YouTube video as JSON, plain text, SRT, or VTT via the TranscriptOut API.

An output-style skill that reshapes replies into a skimmable briefing: TL;DR, status checklist, and numbered choices.

Turns Claude into a cofounder-style brainstorming partner that writes no code, challenges your assumptions, and ends with a structured Brainstorm Brief.

Pulls AI industry news from the smol.ai RSS feed and turns it into a date-scoped Markdown digest, an optional themed HTML page, and a shareable card image.

A short two-batch interview plus 3–5 reference frames that produces brand.md, MOTION.md and a read-first rule in CLAUDE.md so every animation stays on-brand.

A thinking-advisor skill that applies Musk's five mental models and eight decision heuristics to tear down cost structures, test physical feasibility, and design aggressive iteration paths.

A startup branding skill that generates name candidates, trademark/domain/handle checklists, brand voice, and a colour-type-logo brief.

Turns a conversational bug report into a durable GitHub issue written in the project's own domain language.

A thinking-partner skill that applies Charlie Munger's latticework of mental models and inversion to stress-test your decisions.

Finds every CLAUDE.md in your repo, grades it against a quality rubric, and applies only the targeted improvements you approve.

Asks 3-4 targeted questions about your video needs, then recommends the best 1-3 of 16 AI video models with clear reasoning.

A persona skill that forces the laziest solution that actually works — stdlib and native features first, shortest diff, no unrequested abstractions.

Forces every code review comment into one line — location, problem, fix — ready to paste into a pull request.

Sorts pasted LinkedIn DMs and connection requests into leads, recruiters, peers, asks and spam, then drafts only the replies worth sending.