cc-plugin

This skill should be used when the user asks to "start/stop the pool", "toggle the remote-control pool", "keep this project's pool alive", "풀 호스트 띄워줘/내려줘", or to start, stop, or toggle a self-restarting `claude remote-control --spawn worktree` pool host for a project. One singleton per project, tmux-tracked, reachable from the Claude app.

This skill should be used when the user asks to "run codex", "use codex CLI", "delegate to codex", "codex resume", or "continue with codex". Executes tasks via OpenAI Codex CLI with model selection, reasoning effort configuration, and session management.

This skill should be used when the user asks to "download video", "download reel", "save video from URL", "yt-dlp download", "download Instagram reel", "download YouTube video", "download TikTok", "extract audio from video", "download audio from URL", "extract video from QR code", "download media", or provides a URL/QR image from a supported video platform. Handles link extraction, compatibility check, download, and metadata saving via yt-dlp.

This skill should be used when the user asks to "explore claude code source", "find internal features", "investigate cli.js", "check beta headers", "discover hidden settings", or mentions "minified code analysis", "anthropic-beta headers", "context management internals".

Read and interpret Apple Pencil handwriting from Apple Notes (including iPad Quick Note popups) by scanning the local Notes database and reading the rendered drawing images multimodally. Use whenever the user mentions handwritten notes, Apple Notes, Quick Note, Apple Pencil, 노트 필기, 손글씨, 아이패드 메모, or asks "what did I write/jot down", "새 필기 있어?", "노트 봐줘" — even if they don't name the app. Also the substrate for any scheduled polling loop that consumes new handwriting.

This skill should be used when the user asks to "split PDF by chapters", "divide book into chapters", "extract chapters from PDF", or "break PDF into sections". Splits PDF documents based on table of contents or text patterns using pypdf.

This skill should be used when the user asks to "analyze video", "summarize video", "extract video transcript", "understand video content", "video to text", "describe video", "ask questions about video", or "what happens in this video". Analyzes videos using Google Gemini API with local file upload or YouTube URL input.

This skill should be used when the user asks to "unfold" the current work picture or hits one of six recurring moments in multi-PR / multi-project work tracked in Linear: "전체 그림" / "지금 상태 펼쳐줘" / "whole picture" (span-open orient), "다음 뭐 하지" / "뭐가 unblocked" / "next action" (next), "배포 순서" / "머지 전 확인" / "deploy order" (deploy), "이 길로 가는 이유 기록" / "결정 남겨줘" / "log this decision" (decide), "세션 정리" / "구조 변경 반영" / "wrap up the span" (close), "로드맵" / "게이트 어디까지 왔지" / "roadmap" (roadmap). Reads auto-derived state from Linear via MCP; writes ONLY structure and decisions (never status). Invoked as /unfold [moment] [project].

This skill should be used when the user asks to "take browser screenshot", "list browser tabs", "click page element", "navigate browser", "automate browser", "inspect accessibility tree", "monitor network", "run JavaScript in browser", "fill form in browser", "debug web page", or mentions CDP. Attaches to a running CDP instance for browser automation.

This skill should be used when the user asks to "spawn a remote-control session", "open this repo/folder in the Claude app", "remote control here", "띄워줘", "이 디렉터리에서 remote-control 켜줘", or to list/kill those sessions. It launches `claude --remote-control` in a chosen directory as a backgrounded tmux session (`rc-<name>`) reachable from claude.ai/code and the mobile app — no Telegram bridge.

This skill should be used when the user asks to "make a math animation", "animate an equation", "manim animation", "render a Manim scene", "parametric curve animation", "3D plot animation", "animated graph/diagram for a web page", "concept explainer video", or wants a short mathematical / diagrammatic motion clip to embed in an HTML page. Authors a Manim Scene from the request, renders it headless, and returns the video path plus a ready-to-paste `<video>`/`<img>` embed snippet. Default output is a muted, loopable mp4 (no audio track). Out of scope: editing existing footage, audio, or non-math UI/architecture diagrams.

This skill should be used when the user asks to "start dictation", "stop dictation", "toggle voice dictation", "받아쓰기 켜", "받아쓰기 꺼", "음성 받아쓰기 토글", or mentions starting/stopping push-to-talk voice dictation. Toggles the push-to-talk whisper.cpp dictation daemon as a tracked background process.

This skill should be used when the user asks to "find codex session", "codex session context", "lookup codex session", "show codex history", "codex session summary", or provides a Codex session UUID. Finds and summarizes OpenAI Codex session files from ~/.codex/sessions/.

This skill should be used when the user asks to "check hourly updates", "hourly digest", "slack hourly summary", "gmail daily summary", "what happened in the last hour", "hourly monitoring", or needs a periodic digest from Slack or Gmail. Runs as two independent loops on different cadences: Slack hourly, Gmail daily. Also triggers for automated invocation via /loop 1h /hourly-digest:hourly-digest slack and /loop 1d /hourly-digest:hourly-digest gmail. Distinct from per-channel Slack digest or single-source summaries. GitHub/Linear activity sync is out of scope for this skill.

Generate a self-contained HTML slide deck from organized notes. Use whenever the user wants to turn notes, a doc, or session content into a presentation, slides, or a deck. The skill surfaces the audience/difficulty/density/time/scope dimensions progressively (never an all-at-once panel), fills a bundled layout-catalog skeleton (10 layouts, 4 narrative ZONEs, clawd mascot decorators), derives slide count from talk length, and outputs one portable HTML file with assets inlined.

Use when the user wants to spin up, run, start, or self-host a local Excalidraw whiteboard server from the official open-source repo (clone + Vite dev server, no Docker). Handles cloning, the yarn-classic toolchain, a configurable port, port-collision detection, and readiness verification.

This skill should be used when the user wants a directed graph — a DAG, workflow, pipeline, or dependency tree — drawn as a plain-text terminal picture: "draw this graph", "sketch the DAG", "diagram these dependencies", "render as ASCII", or turning a Workflow script's pipeline()/parallel() structure into a diagram (even when "ASCII" is never said). Renders layered box-art, zero-dependency, upgrading to graph-easy when installed.

This skill should be used when the user asks to "start clawd", "stop clawd", "toggle clawd", "launch clawd on desk", "kill clawd", "run the desktop pet", or mentions starting/stopping the clawd-on-desk Electron app. Toggles `npm start` for clawd-on-desk as a tracked background process.

This skill should be used when the user explicitly asks to "bootstrap CDP on Linux", "start Chromium for cdp-attach on a headless box", "set up Xvfb for CDP", or "make cdp-attach work in this sandbox". Launches Playwright-bundled Chromium under Xvfb with --remote-debugging-port so the existing cdp-attach skill passes its headed-browser guard. Linux-only; user-invoked only (does not auto-trigger from cdp-attach errors).

This skill should be used when the user asks to "remove program", "uninstall app", "delete daemon", "clean up process", "safe uninstall", "find resource hog", "what is this process", "remove bloatware", "kill daemon", "remove LaunchDaemon", "remove security program", "Activity Monitor finding", "high memory process", or wants to safely identify and remove macOS programs and daemons. Also triggers when the user mentions high memory/CPU usage by unknown processes, Activity Monitor concerns, unwanted LaunchDaemons/Agents, or Korean banking security programs (TouchEn, AhnLab, Veraport, INISAFE, CrossEX). Use this skill even when the user simply provides a process name and wants it gone, or describes a process consuming excessive resources they want to investigate.

This skill should be used when the user asks to "prevent sleep", "caffeinate", "keep awake", "stop sleeping", "clamshell mode sleep prevention", "toggle caffeinate", or mentions system sleep prevention on macOS. Toggles caffeinate -ims (idle, disk, system sleep prevention; display sleep allowed for battery saving).