Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Crafting puzzles that teach through play, create "aha" moments, and challenge without frustrating. Use when "puzzle, puzzle design, puzzle game, puzzle mechanics, difficulty curve, hint system, brain teaser, riddle, logic puzzle, escape room, teaching through play, aha moment, portal puzzle, baba is you, the witness, talos principle, professor layton, puzzles, game-design, level-design, difficulty, teaching, hints, brain-teaser, escape-room, logic, challenge" mentioned.
Creates programmatic video with React via Remotion — compositions, sequences, and motion graphics animated with useCurrentFrame() and rendered to MP4. USE WHEN video, animation, motion graphics, video rendering, React video, render video, animate content, make a short, create animations, video overlay, explainer video, animated explainer, content to video, programmatic video. NOT FOR static images, diagrams, or illustrations (use Art), tight-cut filler removal from raw talking-head recordings (that is video editing, not motion graphics), or audio-only podcast cleaning (use AudioEditor).
Guidance for building programmatic videos with rendiv — a React/TypeScript framework for composing video scenes, animating with springs and interpolation, and rendering to MP4/WebM. Use when writing or modifying rendiv compositions, working with rendiv animation APIs, or setting up a rendiv project.
Provider-independent workflow for producing SaaS and software product demo videos, including app walkthroughs, feature launches, sales demos, onboarding clips, help-center explainers, website hero demos, and PLG assets. Use when an agent must plan, script, capture, synthesize, edit, localize, review, or QA privacy-safe software demo videos with substantiated claims, accessible captions, screen/cursor direction, CTA variants, and stakeholder approval.
Generates a new image that imitates the style of a reference image while updating content based on user intent. Uses a three-stage pipeline: image annotation (long caption), caption rewriting, and image generation. Use when user asks to "imitate style", "保持这个风格重画", "按这张图风格生成", or "style transfer with new content".
Create and manage Spotify playlists, search music, and control playback using the Spotify Web API. UNIQUE FEATURE - Generate custom cover art images (Claude cannot generate images natively, but this skill can create SVG-based cover art for playlists). CRITICAL - When generating cover art, ALWAYS read references/COVER_ART_LLM_GUIDE.md FIRST for complete execution instructions. Use this to directly create playlists by artist/theme/lyrics, add tracks, search for music, and manage the user's Spotify account.
Intelligently create and publish WeChat Official Account articles. Supports AI article generation, link publishing, Markdown conversion, image cleanup, style processing, and one-click publishing. Use when the user mentions "创作公众号文章", "发布文章链接", "生成微信文章", "推送到微信公众号", "发布到微信公众号".
Manage Alibaba Cloud ApsaraVideo for Media Processing (MPS/MTS) resources and workflows via OpenAPI/SDK. Use for media ingest and metadata tasks, transcoding/snapshot jobs, pipeline/template/workflow operations, and MPS job troubleshooting.
Bake texture maps in Blender 5.2 Cycles — high-to-low normal maps with cage/extrusion, AO, curvature, ID, thickness, base color/roughness/metallic from procedurals, ORM packing, OpenGL vs DirectX, margins, GPU baking, and fixing skewing/waviness/seams. Use when transferring sculpt or high-poly detail to game meshes or converting procedural materials to textures
Use when the user wants to "make a video/clip", "generate a video from text", "animate this image/photo", "create a marketing reel / social video / teaser", "do a 3D / parallax move", "pan/zoom/orbit a shot", or mentions fal.ai, Kling, Veo or Seedance for video. Staged fal.ai workflow — brainstorm, draft cheaply, produce the final at quality. Three modes: text-to-video, animate (still to motion), camera-move (push-in/pan/orbit). Not for still images (image-studio), slide decks, or charts. Honesty discipline: never depict real identifiable subjects via text-to-video.
Orchestrate the full workflow for analyzing plot points and dramatic functions in long texts: preprocessing, segmented analysis, and result integration. Use when a novel or script needs a structured dramatic analysis report, or when a dramatic workflow or plot-point analysis pipeline is mentioned.
Remove signs of AI-generated writing from text. Use when editing or reviewing text to sound natural and human-written. Based on Wikipedia's 'Signs of AI writing' — catches em dash overuse, rule of three, AI vocabulary, vague attributions, inflated symbolism, negative parallelisms, passive voice. Not for agent-facing text.
A tool for tracking hit short-drama content on Bilibili. It automatically scans Bilibili short-drama content daily, filters hit works by like count, intelligently clusters topics, and generates a visualized HTML daily report with creation trend analysis. ⚠️ Before querying, the script validates input first: the keyword must hit the short-drama topic word list (the topic name plus all related words defined in topic_keywords, e.g. 「打脸」 hits related words of the counterattack topic); once matched, the keyword is used directly to query data; if not matched, it reminds '关键词不满足查询条件' and recommends related words, and does not issue an API request. Data is updated daily at 15:00 for the previous day's data; when there is no data for the target date, you must first inform the user and wait for confirmation before calling the API, automatic fetching is prohibited. Use when the user needs to query Bilibili short-drama hit daily reports, analyze short-drama topic trends, view popular creator performance, or generate short-drama creation trend reports.
Short drama – Xiaohongshu information source: scans hit short-drama content on Xiaohongshu daily, filters popular notes by engagement, intelligently clusters topic directions, and generates an HTML daily report containing covers, engagement data and creation insights. Supports targeted queries by topic (time travel/domineering CEO/rebirth, etc.), creator, and time range. ⚠️ Data is updated daily at 15:00 for the previous day's data; when there is no data for the target date, you must first inform the user and wait for confirmation before calling the API, automatic fetching is prohibited. ⚠️ Before querying, first validate whether the category/keyword matches the short-drama topic word list; if not, do not request the API, directly notify and recommend related words. Use when the user needs short-drama Xiaohongshu daily reports, Xiaohongshu short-drama hits, short-drama hot topics, short-drama creation trends, or custom topic queries.
Diagnose, design, script, produce, adapt, and quality-control product video advertisements from evidence-backed briefs. Use for physical-product, software, app, service, ecommerce, direct-response, brand-response, demonstration, testimonial, launch, and variant ad work where product truth, legibility, claims, platform delivery, and measurable iteration matter.
When the user wants editing guidance for YouTube retention, pattern interrupt techniques, pacing direction, or re-engagement strategies. Also use when the user says 'editing for retention,' 'pattern interrupts,' 'how should I edit this,' 'pacing guide,' 'my videos lose viewers in the middle,' 'editing notes,' 'retention editing,' 'keep viewers watching.' For script-level retention, see script-structure. For analyzing retention data, see video-analysis.
Comprehensive analysis of the five story elements: genre, synopsis, character bios, character relationships, and major plot points. Suitable for in-depth story analysis, screenplay adaptation preparation, and story development documentation.
Japanese (and other non-Hollywood) screenwriting methods (日本编剧方法/原创剧本实战) — distilled from ten Japanese directors and writers in ゼロからの脚本術; Kosawa Ryota (three acts as 起承转合, speed by jumping), Uchida Kenji (beautiful structure, time-axis tricks), Miki Satoshi (small-material notebook, theme emerges later), Sono Sion, Omiya Ellie (dialogue from inner monologue), Kakei Masaya ("if + moreover"), Fukuda Yuichi (character = actor + flaw, audience as tsukkomi), Yokohama Satoko, Takahashi Izumi ("pain"), Yukisada Isao (situation-born originals); plus Arai Haruhiko (IP adaptation, "write what you're ashamed of"). Neighbours; sw-korean-french-screenwriting, ozu-screenplay-style. Use when writing an original Japanese-style script or TV drama, working from small material and fragments, choosing between structure-first and character-first methods, building a Japanese 企画書, designing comedy tsukkomi systems, or when Hollywood models feel wrong for a quiet, daily-life, non-goal-driven story.
Unity shaders, materials, and rendering pipelines (URP/HDRP/Built-in). PROACTIVELY activate for: (1) writing shaders in Shader Graph, HLSL, or ShaderLab, (2) URP and HDRP shader authoring, (3) custom render pipeline work (SRP), (4) lighting setup (baked vs realtime, lightmaps, Global Illumination), (5) post-processing stacks, (6) reflection probes and light probes, (7) custom render features and full-screen passes, (8) shader stripping and variant management, (9) compute shaders, (10) ray tracing in HDRP. Provides: Shader Graph templates, HLSL snippets, URP/HDRP differences, lighting setup recipes, render-feature examples, and shader-variant guidance.
After you enter a link from any video site or podcast platform, the corresponding media file is downloaded automatically and delivered to the user. yt-dlp is preferred, covering common video sites such as Douyin, Bilibili, and YouTube, and it can also be used for podcast platforms that expose a direct audio URL (such as a single-episode link from Xiaoyuzhou). When 403 errors, login, age, or region restrictions occur, a retry with cookies.txt is supported; for platforms that may have DRM/encryption or terms restrictions (for example some Spotify content), prompt the user to download only content they have the right to save, and when download is not possible, suggest official offline/export channels or provide the original RSS/direct link. When a Douyin video hits yt-dlp's signature wall ("Fresh cookies needed"), it automatically falls back in three stages: a direct unauthenticated connection to aweme.snssdk.com/v1/play → re-running yt-dlp with local browser guest cookies (no Douyin login needed), all seamlessly; Douyin image-text notes are not yet supported for automatic download and require manual handling.
Add impact and feedback effects to a Three.js game with lightweight particle bursts — muzzle flashes, hit sparks, blood/dust, explosions, pickups — using pooled additive sprites or Points. Use when actions need visible punch (firing, impacts, deaths, explosions) or for ambient effects (smoke, embers, dust).
Quickly generate 2-3 video script outline options, including title suggestions, thumbnail design suggestions and a complete structural design. Use this skill when the user mentions "video outline", "video script", "video planning", "shooting a video", or "video content".
Render with Blender 5.2 Cycles/EEVEE/Workbench - quality presets, GPU setup, adaptive sampling, denoising, light paths, color management (AgX, Khronos PBR Neutral, ACES), PNG/EXR/MP4 output, passes, the 5.0 compositor API, CLI and multi-camera batch rendering. Use for any still, animation or turntable render.
Set up and audit Blender scenes the way studios do - units and real-world scale, collection layout, naming conventions (SM_/SK_/T_/M_), origins, linking vs appending, library overrides, asset libraries and catalogs, collection exporters, orphan/path cleanup. Use when starting a project, organizing a .blend, or before handoff/export.
Multimodal video understanding: generate descriptions, summaries, and tags for videos, and answer questions based on video content. Trigger when the user wants to understand video content, generate descriptions, or ask questions about a video.
Capture frames from a YouTube video at a regular interval, produce a manifest mapping filenames to timestamps, and describe each frame with an LLM. USE FOR: extract video frames, capture screenshots from YouTube, describe video frames, video frame analysis, frame-by-frame summary.
All-in-one TikTok Shop selling-video skill for any AI agent (Claude Code, Codex, WorkBuddy, OpenClaw). Find viral TikTok videos, research TikTok Shop products, shops, creators and live rooms, break down why a video sells (script, scenes, hooks, music), search the largest library of real high-GMV AI selling videos and their reverse-engineered prompts and adapt the closest match to your own product, replicate a winning video, turn product photos — or your own raw prompt and images — into AI selling / UGC / talking-head / product-demo videos, generate e-commerce images from a text prompt, upscale to 1080p or 2K, and download TikTok or Douyin videos. Keywords — AI selling video, TikTok viral replication, TikTok Shop product research, competitor shop analysis, creator and influencer ranking, AI selling video prompt library, product-to-video, UGC video generator, AI product image, TikTok video downloader. Use whenever the user needs TikTok e-commerce data, viral video research, or AI video/image generation.
Content atomization — turn one piece of content into many formats. Covers blog-to-thread, blog-to-carousel, podcast-to-blog, video-to-quotes, and more. Use for: content marketing, social media, multi-platform distribution, content strategy. Triggers: content repurposing, repurpose content, content atomization, content recycling, one to many content, multi platform content, cross post, adapt content, reformat content, blog to thread, blog to video, podcast to blog, content multiplication
Invoke when the user asks about story structure, branching narrative, dialogue systems, world-building, character design, environmental storytelling, ludonarrative coherence, writing for games, lore, theme, or character arcs. Triggers on: "story", "narrative", "dialogue", "lore", "world-building", "character arc", "branching", "ludonarrative". Do NOT invoke for creative vision (use game-creative-director) or game mechanics (use game-designer). Part of the AlterLab GameForge collection.
Provider-independent production workflow for AI agents creating game trailers and promos, including announcement, gameplay, launch, DLC/update, store-page videos, social cutdowns, and generated/engine-capture hybrid trailers. Use when planning, scripting, capturing, generating, editing, localizing, reviewing, or QAing game marketing videos where gameplay truthfulness, platform/store constraints, ratings, accessibility, claims, CTA metadata, and rights provenance matter.
Beginner-friendly end-to-end video creator for HyperFrames. Use when the user says "make a video", "create a video", "new video", "build a video", "video from scratch", "I want to make a video", "help me create a video", or when someone who's never used HyperFrames before arrives with a concept, script, or rough idea and wants a finished MP4. Interviews the user in one pass, then builds the full video with mandatory preview and visual-verification gates.
AI video director pipeline: automatic script storyboarding (with WenWu director-level shot transitions and character four-view reference images) → MiniMax H3 workflow allocation and parameter confirmation → batch generation monitoring → automatic CapCut assembly into a finished film, with optional Cloudflare tunnel remote access. Prompts support the official H3 six-part format (official), the WenWu Chinese director mode (wenwu), and a hybrid mode that combines both (official six-part shell + WenWu director-level content). Use when the user mentions script-to-video, automatic storyboarding, shot transitions, WenWu, character four-view, H3 generation, workflow allocation, CapCut assembly, AI director, game PV/promos/ad shorts/style and music briefs, or provides a script/story text and asks to generate a video.
Use when a 2D side-scroller shot, platformer cutscene, game trailer or parallax animation has to be built in Blender from Scenario-generated pictures (flat image cards, an orthographic camera, keyed jumps, rigid-body props), or when a finished shot needs a new theme or look without re-animating. Also use when image planes in Blender float above the ground, a tiled parallax layer shows a seam, or physics props leave the picture plane.
YouTube thumbnails, video/podcast/course covers, blog headers, and Open Graph (link-preview) images via Gemini 3.1 Flash image generation (default) or HTML/CSS. Optimized for high contrast, small-size readability, and click-through rate. Use when the image represents a video, episode, or article; for feed posts use social-media-graphic, for paid ads use ad-creative-design. Trigger phrases: "design a thumbnail", "YouTube thumbnail", "video thumbnail", "article header image", "OG image", "create a thumbnail", "blog header", "video cover image".
Use meitu-cli for comprehensive video quality restoration (deblurring/denoising/frame interpolation/brightening shadows), auto-adapting to portraits, concerts, anime, night scenes, and products. Trigger when the user mentions 画质修复, 视频修复, 变清晰, 老视频翻新, or one-click video enhancement without a resolution requirement.
Build multi-step AI content creation pipelines combining image, video, audio, and text. Workflow examples: generate image -> animate -> add voiceover -> merge with music. Tools: FLUX, Veo, Kokoro TTS, OmniHuman, media merger, upscaling. Use for: YouTube videos, social media content, marketing materials, automated content. Triggers: content pipeline, ai workflow, content creation, multi-step ai, content automation, ai video workflow, generate and edit, ai content factory, automated content creation, ai production pipeline, media pipeline, content at scale
Create AI marketing videos for ads, promos, product launches, and brand content. Models: Veo, Seedance, Wan, FLUX for visuals, Kokoro for voiceover. Types: product demos, testimonials, explainers, social ads, brand videos. Use for: Facebook ads, YouTube ads, product launches, brand awareness. Triggers: marketing video, ad video, promo video, commercial, brand video, product video, explainer video, ad creative, video ad, facebook ad video, youtube ad, instagram ad, tiktok ad, promotional video, launch video
Lighting, cameras, and look development in Blender 5.2 — studio three-point and product lighting, HDRI environments, sun/sky exteriors, interiors, light linking, physically based light units, camera focal lengths/composition/DOF, cyclorama backdrops, and looping turntables for asset presentation. Use when presenting assets, setting up scenes for rendering, or judging materials under consistent light
Unified pipeline for content production and distribution. Material collection → drafting → layout → cover → Moments copy → multi-platform conversion → one-click distribution. Covers WeChat official account writing, Xiaohongshu carousel images, Jike copy, podcast audio, brand videos, and Chrome CDP automated publishing.
FrameVideo's Chinese AI production sub-workflow. Starting from a raw script, plot synopsis, existing storyboard Clip, video Shot, reference image asset, or single-video idea, it handles script-to-storyboard conversion, aggregating storyboards into video tasks, and prompt polishing, and when real generation is needed, hands off to the Chanjing 蝉镜 AIGC adapter to generate image/video assets and backfill the FrameVideo composition. Use for scenarios where the user requests end-to-end AI video production, video workflow orchestration, storyboard script generation, video task planning, Clip aggregation into Shot, Chanjing AIGC image/video generation task planning, or a path from idea through to FrameVideo preview/render.
Create PowerPoint/PDF presentations where each slide is a complete AI-generated image with all text embedded. Supports multiple visual styles (corporate, minimalist, creative, kawaii, custom). Use when user wants fully-illustrated slides, visual-first decks, or AI-generated presentation images via x402 payments.
Use when the user wants image generation or editing through the local fal.ai three-stage studio workflow — brainstorm a strong prompt, prototype cheaply and iterate on feedback, then produce a finalised image. Covers "generate an image", "make an image/picture/illustration/graphic of…", "create an AI image", "edit/change this image", "make a variation", "upscale this", a "production-ready image", "clean up / enhance a photo", "make this phone shot look professional", imagery "for the deck/post/website/newsletter", or explicit mentions of fal.ai or nano-banana. Do not use for Canva template designs, branded PowerPoint decks, data charts/dashboards, or flowcharts/diagrams. Video generation is out of scope.
NotebookLM integration patterns for external RAG, research synthesis, studio content generation (audio, cinematic video, slides, infographics, mind maps), and knowledge management. Use when creating notebooks, adding sources, generating audio/video, or querying NotebookLM via MCP.
Write natural, human-sounding Persian (Farsi) and build correct right-to-left Persian deliverables. Use for ANY task involving Persian text or an Iranian audience: writing, editing, translating, humanizing, SEO content and copywriting (سئو، کپیرایتینگ، لندینگ), academic papers/theses, and Persian output in Word/.docx, PDF, PowerPoint, HTML, Excel, images/posters, emails, social posts. Covers register detection (رسمی/اداری/محاوره/علمی — errors here are costly), Persian AI-tell removal, orthography (نیمفاصله/ZWNJ، اعداد فارسی، «گیومه»), cleanup and spell-check (ویرایش، پاکسازی — bundled paknevis+davat toolkit), RTL layout/pagination fixes, and Persian fonts (Vazirmatn bundled). Trigger whenever you see Persian/Arabic script or words like فارسی, Farsi, Persian, Iran, RTL, راستچین, نیمفاصله, ویرایش, مقاله, پایاننامه, سئو, کپشن, Vazirmatn — even if the user never mentions this skill.
Run any model on RunComfy from the command line. The `runcomfy` CLI is one binary, one auth, hundreds of model endpoints — image generation, image edit, video generation, image-to-video, lip-sync, face swap, video edit, inpainting, outpainting, extend, ControlNet, relight, upscale, LoRA training and more. Submit a request, poll for status, download the output. This skill teaches the agent how to install, authenticate, discover model schemas, invoke models, stream / poll / no-wait, script in JSON output mode, and handle errors. Triggers on "runcomfy cli", "install runcomfy", "runcomfy login", "runcomfy run", "runcomfy whoami", "runcomfy api", or any explicit ask to call a RunComfy model from a script or terminal. Sibling skills (ai-image-generation, ai-video-generation, image-edit, video-edit, face-swap, lipsync, image-to-video, image-inpainting, image-outpainting, video-extend, controlnet-pose, relight) all dispatch through this CLI.
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.