Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
After completing any task that took more than ~5 tool calls, or after long-running builds/tests finish, load this skill and deliver a dad joke to lighten the mood. Also load before any user-requested joke, pun, or limerick. Never improvise jokes without loading this skill first.
Use when the user asks you to write, draft, rewrite, or edit prose for them to send or publish: Slack messages, PR descriptions, tickets, docs, emails, release notes, blog posts, commit bodies, or a summary or plan they asked for as the deliverable. Strips the LLM-essay register (em dashes, antithesis, corrective negation, rule of three, setup/payoff, throat-clearing openers, landing sentences, nominalization, hedging, performed enthusiasm, metaphor nouns, passive voice, adverbs) and writes for the spoken voice instead. Also use when the user says "no em dashes", "sounds like AI", "make it sound human", "strip the AI voice", or invokes /dispel. Do not load it for your own routine summary or plan at the end of other work
Keep technical nonfiction durable by separating timeless concepts, decision frameworks, and reader outcomes from fast-changing tool walkthroughs, UI steps, versions, vendor details, and companion resources. Use when scoping, outlining, revising, or updating technical books, self-hosting guides, software manuals, DevOps books, or other technical manuscripts that risk becoming outdated.
Plan, source, vet, and precisely place B-roll for spoken video. Use when the user asks to find B-roll, source cutaways, add receipts or visual evidence over a talking head, illustrate a transcript, create an annotated B-roll plan, or time cutaways to exact spoken words. Handles official footage, product interfaces, public-page receipts, archival material, concept graphics, and user-curated cultural clips while preserving source provenance and final editorial approval.
Use when the user shares an audio file (mp3, wav, m4a, flac) and asks to transcribe it to text — includes queries containing "speech recognition", "audio to text", "transcribe", "transcription", "ASR", "meeting notes", "convert audio", "subtitle", "SRT", "speaker diarization", "who said what". Also triggers on audio meeting recordings, interviews, phone calls, lectures, voice memos, podcasts. Requires DASHSCOPE_API_KEY (Alibaba Cloud DashScope) and S3-compatible storage credentials.
Design, implement, budget, and verify readable real-time game visual effects: particles, smoke/fog, lightning, fire, trails, impacts, bloom/glow, and layered spell timelines. Use when a user needs a VFX spec, lifecycle, engine handoff, particle pooling, effect composition, reduce-motion fallback, or frame-budget diagnosis. Triggers on: game VFX, spell effects, particle system, impact burst, procedural lightning, smoke, bloom, magic aura, or VFX performance.
Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.
Image resizing and compression tool skill. Used to resize images by specified pixel width and height, ratio, or maximum size limit, and supports smart compression to a specified file size. Applies to scenarios such as batch image processing, generating thumbnails at specific sizes, and compressing images to meet file size limits.
LibTV (LibLib.tv) advanced video workflow skill: one-click AI video generation, automatic download, and smart editing (with transitions, subtitles, BGM). Use when the user wants to create AI videos or short films using LibTV, especially when they need automatic video editing with transitions, subtitles, or background music.
Turn a long video into N viral-ready short clips with a single managed API call. Wraps muapi.ai's `/ai-clipping` endpoint, which handles transcription, highlight ranking through a virality framework (hook / emotional peak / opinion bomb / revelation / conflict / quotable / story peak / practical value), overlap dedupe, and vertical face-tracking auto-crop server-side. No local Whisper, no local LLM, no GPU.
Expert Cinema Director skill for Seedance 2.0 (ByteDance) — high-fidelity video generation across Chinese, Global, and VIP tiers. Supports text-to-video, image-to-video, first-last-frame, omni reference, character training, omni-reference training, video editing, and watermark removal.
Auto-generate viral 9:16 YouTube Shorts (or TikTok / Reels clips) from a long-form video. Thin platform-aware wrapper around the AI Clipping skill — picks sensible defaults for short-form social platforms (9:16, 30–60s sweet spot) and delegates the actual highlight extraction + crop to muapi.ai's `/ai-clipping` endpoint.
Generate music using ElevenLabs Music API. Use when creating instrumental tracks, songs with lyrics, background music, jingles, or any AI-generated music composition. Supports prompt-based generation, composition plans for granular control, and detailed output with metadata.
Train custom TTS voices for Piper (ONNX format) using fine-tuning or from-scratch approaches. Use when creating new synthetic voices, fine-tuning existing Piper checkpoints, preparing audio datasets for TTS training, or deploying voice models to devices like Raspberry Pi or Home Assistant. Covers dataset preparation, Whisper-based validation, training configuration, and ONNX export.
Master-control skill pack for product promo video production. Used to produce 60-90 second promo videos from product descriptions, official sites, app screenshots, or GitHub repositories, completing the brief, storyboard, assets, HyperFrames editing, BGM design, and delivery in stages. Use when the user wants to make a promo video, product video, project intro video, launch video, open source project promo, BGM beat-synced cut, or promo video featuring real software interfaces.
Use this skill to record a terminal or TUI demo with VHS whenever the user wants to "record a demo", "create a GIF of my CLI", "write a VHS tape", "make a terminal recording", or "add a demo GIF to the README" — even if they never say "VHS" and only ask for an animated demo of a command-line tool. Covers tape syntax, dimensions, pacing, GIF size reduction, and CI golden-file recording. Do NOT use this skill to build the TUI itself (see build-tui), to screenshot a web app on localhost (see screenshot-local), or to capture video of anything outside a terminal.
Editing Russian-language texts using Lyudmila Sarycheva's method — from meaning to words, not the other way around. Use this skill whenever the user asks to edit, review, improve, shorten, or polish any Russian-language text — articles, posts, emails, landing copy, docs. Triggers: «отредачь», «отредактируй», «почисти текст», «сократи», «перепиши по-человечески», «убери канцелярит», «сделай проще», «причешите текст», «дай фидбек на текст», «разбери мой текст», «слишком водянисто», «звучит как робот». Also trigger when the user pastes a Russian draft and asks what's wrong with it or how to make it better. For writing text from scratch, see skill statya; for interface texts — ux-copy.
Create polished, code-driven Remotion motion explainers from scripts, narration, subtitles, product specifications, reference videos, UI recordings, diagrams, or abstract concepts. Use for product launches, AI demonstrations, UI interactions, algorithms, scientific principles, electrical systems, data flows, architecture, charts, equations, and concepts best expressed with React, SVG, Canvas, WebGL, or Three.js. Prioritize continuous spatial logic, visible cause and effect, precise interaction states, editable parameters, deterministic rendering, and evidence-backed visual QA.
Use when making or fixing Roblox visual effects or game audio: ParticleEmitter, Beam, Trail, flipbook sprite sheets, slash, hit, impact, explosion, fireball, aura or projectile VFX; effects only one player sees, effects that lag or pile up, particles invisible, popping or too bright; the Audio API (AudioPlayer, Wire, AudioEmitter, AudioListener, AudioDeviceOutput, effects), "no sound", "3D sound not working", music, ducking, footsteps, SoundGroup volumes; or importing Scenario-generated sound effects, music, voice or VFX sheets into a Roblox game.
Transcribe a ScreenKite recording's mic audio, then remove BOTH silence gaps AND filler words (um, uh, ah, er, hmm) from the timeline in a single merged editTimeline cut. Use when the user asks to "clean up audio", "remove filler words", "remove ums and uhs", "cut dead air and fillers", or "full auto-cut" on a .skbundle project. Combines screenkite-transcription-cut silence removal with filler-word detection into one dry-run → confirm → apply flow. Do NOT use for B-roll overlay work (see use-screenkite-advanced-b-roll). Do NOT use when no .skbundle is open.
Writes, debugs, and art-directs procedural GLSL fragment shaders (Shadertoy mainImage, WebGL2 / GLSL ES 3.00): SDF raymarching, procedural buildings/cities, 2D distance-field graphics, noise / fBm / domain warping, terrain, ocean/coastal water optics, clouds and other volumes, PBR lighting, color and tonemapping. Ships a verified GLSL building-block library with a bundler and checker. Use when the user asks for a shader, Shadertoy, GLSL/.frag/.glsl code, a raymarcher, an SDF scene, a procedural texture or generative visual, or wants a shader fixed, optimized, ported or explained. Not for HLSL/Unity ShaderLab or non-graphics GPU compute.
TikTok video download — paste a TikTok video link to get a watermark-free video download link in one step, with support for single or batch link parsing and automatic TikTok link validation. Use when the user needs to download TikTok videos, save TikTok videos, get a TikTok video direct link, or batch download TikTok videos. Trigger words: TikTok video download, TK video download, TikTok video parsing, download TikTok video, Douyin international version video download, batch download TikTok videos.
Extract audio from video files or URLs (including YouTube). Supports MP3, WAV, M4A, FLAC, OGG, and OPUS formats. Can process local video files or download from URLs. For YouTube videos, uses yt-dlp for direct audio extraction when possible.
Download videos from Instagram, YouTube, TikTok, Twitter/X, Facebook, and 1000+ other platforms as MP4 files using yt-dlp. Auto-recovers from YouTube bot detection ("Sign in to confirm you're not a bot", LOGIN_REQUIRED) by rotating through browser cookies and cookie files without manual intervention. Use this skill when the user wants to download, save, or grab one or more videos from any social media or video platform URL. Supports single videos or full playlists, configurable resolution, and saving to a specified target folder.
Edit existing videos with AI using each::sense. Apply effects, color grading, speed changes, trimming, transitions, style transfer, and visual enhancements. Transform raw footage into polished content. Use for: color grading, speed ramping, style transfer, video enhancement, social media edits, content post-production. Triggers: edit video, video editing, color grade, speed change, video effects, trim video, video filter, slow motion, timelapse, video style, video enhance, post production
Analyze video files by detecting scene boundaries and generating structured production metadata as JSON. Use when asked to analyze scenes in a video, detect scene changes, generate scene-level metadata (camera angles, color tone, tempo, transitions, text overlays), catalog visual assets, or produce a structured JSON breakdown of video content. Includes bundled videx script for frame extraction (requires ffmpeg/ffprobe). Outputs analysis.json with per-scene content summaries, production/direction tags, and a cross-scene asset library.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.