Skills
96,770 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Use for Blender modelling, materials, scenes, rigging, animation, rendering, inspection, and exports. Research visual references for new or substantially changed artwork, build with reproducible Blender Python, and inspect renders and scene data before delivery.
Removes unwanted visual elements from videos using AI-powered inpainting via Volcengine LAS. Video watermark removal, subtitle removal, logo removal, and text overlay removal — erases and cleans up watermarks, hardcoded subtitles, logos, text, and other fixed-region artifacts and obstructions from video footage. Video inpainting, video repair, video restoration, and video cleanup. Supports specifying exact bounding boxes for targeted removal and erasure. Handles videos up to 4 hours and 30GB. Use this skill when the user wants to remove watermarks from videos, erase hardcoded subtitles/captions, delete logo or text overlays, clean up video footage by removing unwanted objects, repair or restore videos by removing obstructions, do video inpainting on fixed regions, or remove visual artifacts from video content.
Use when the user wants Claude to send images to OpenAI Codex — Codex-side review of an existing screenshot/mockup/wireframe, generate an image via Codex's image_gen tool, or edit an existing image via Codex — or explicitly invokes "codex-vision" / "/codex-vision". Wraps the upstream `codex exec -i` CLI with three modes (review / generate / edit) and an opt-in tmux session prefixed `claude-codex-` so the user can attach and watch. SKIP when the task is text-only, or when the user needs a durable multi-turn repo task — use codex-tmux-dev for long-running image+coding work. The bare word "codex" is not a trigger; the bare word "screenshot" is not a trigger. For generate/edit, require explicit Codex intent unless the user invoked the skill by name. Explicit `/codex-vision` invocation overrides every other gating rule below.
When the user wants to turn one piece of content into many, repurpose a video/podcast/post for other platforms, create social posts from long-form content, or says things like 'repurpose this,' 'turn this into,' 'social posts from,' 'newsletter from this video,' 'clips from this.' Takes one input and multiplies it across platforms.
Generate an animated, interactive HTML page that explains a complex concept through real-world analogies, visual diagrams, and progressive disclosure. Use this whenever the user says 'explainer infographic,' 'explain how X works,' 'make an explainer for,' 'visualize this concept,' 'turn this into a visual,' 'help me understand X visually,' or asks to break down a topic in a bite-sized visual format. Output is a single self-contained HTML file the user can open in a browser, scroll through, or share as a link. Topics can be anything — finance concepts, technical systems, scientific processes, business frameworks.
Create game-ready fixed-cell pixel-art character spritesheets from text concepts, reference images, or existing character art. Use when Codex needs to design, generate, clean up, validate, and package 2D game characters with native 32x32, 64x64, or 128x128 cells; 4-way or 8-way directions; idle, walk/run, jump, attack, archer, caster, or other animation strips; transparent GIF/WebP previews; and per-direction targeted regeneration.
Expert in casting, directing, and generating expressive text-to-speech using Gemini TTS (Gemini 3.8 Flash / Flash-Lite TTS and earlier 3.1 / 2.5 models). Use this when the user needs virtual voice actor personas, expressive speech generation, two-speaker dialogue, or multiple variations of a voiceover (like "take 3 on the bounce")
Elimina patrones de escritura típicos de IA en textos en español de España para que suenen naturales y humanos. Aplica esta skill siempre que generes, edites o revises textos en español: artículos, guías, tutoriales, emails, copy comercial, publicaciones en redes sociales, documentación, informes o cualquier prosa. Actívala también cuando el usuario pida humanizar texto, eliminar tono robótico, mejorar la naturalidad de un texto, o cuando mencione que algo suena a IA, a ChatGPT o a texto generado automáticamente.
Produce a LinkedIn-native 1:1 (or 16:9) launch video using HyperFrames. Caption-first, sound-off, B2B register, anti-corporate-jargon. Optionally hand off to LinkedIn Ads to launch as a sponsored content campaign with ICP-aligned targeting.
Transforms any subject knowledge point into a vertical educational infographic (9:16). The user only needs to enter a knowledge point name (such as "sound phenomena," "Pythagorean theorem," "photosynthesis"), and the Skill automatically: identifies the subject/grade/chapter, breaks it down into 3-5 core sub-concepts, assigns a theme color scheme, generates specific illustration descriptions for each sub-concept, builds a complete prompt, and calls linyuebanzi-image-gen to generate the image. Built-in two-layer subject accuracy check: self-check the prompt against a per-subject checklist before generation, then read back the PNG for visual review after generation (text/formulas/diagrams), with automatic retries on failure to ensure the content delivered to students is accurate. Built-in illustration guidance for subjects including physics (sound/light/heat/force/electricity/lenses/mechanical motion), math, chemistry, biology, Chinese, history, and English. Supports batch generation - "all chapters of first-semester 8th grade physics" will be generated chapter by chapter, each with a different color scheme and a consistent style. Trigger words include: make an infographic, knowledge card, subject diagram, physics infographic, chemistry infographic, make a knowledge chart, draw a knowledge card, turn this knowledge point into an image. Do not use for dynamic interactive demos (use the interactive demo Skill), article cover images (use linyuebanzi-cover-generator), or hand-drawn style article illustrations (use linyuebanzi-inline-diagram).
Explain anything — turn ideas into podcasts, explainer videos, or voice narration. Use when the user wants to "make a podcast", "create an explainer video", "read this aloud", "generate an image", or share knowledge in audio/visual form. Supports: topic descriptions, YouTube links, article URLs, plain text, and image prompts.
Literary ghostwriter skill that mimics the writing style of seven literary masters for creative writing. Extended with long-form novel project management, multi-role writing workflow, and consistency tracking. Triggers when: Writing in a specific author's style, creative writing, script writing, novel writing, long-form fiction planning, chapter continuation, or style imitation exercises. Commands: - /文豪 <作家> <选题> — Single-shot style imitation (7 authors) - /literary <author> <topic> — English single-shot style imitation - /文豪 新建 <项目名> — Create novel project with world-building - /文豪 打开 <项目名> — Open existing project - /文豪 项目 — Show project status - /文豪 规划 — Analyze novel position, plan next moves - /文豪 大纲 [操作] — Generate/modify plot outline - /文豪 章纲 <章节号> — Generate chapter-level outline - /文豪 设定 <世界观元素> — View/modify world-building - /文豪 人物 [角色名] — View/modify character card - /文豪 伏笔 — View foreshadowing tracker - /文豪 续写 — Smart continuation from current position - /文豪 润色 <文本> — Style optimization - /文豪 审稿 [章节] — Editor review - /文豪 去AI味 <文本> — AI-trace reduction - /文豪 回收站 — Recycle bin for deleted content Capabilities: Deep analysis of author's spiritual core, providing concrete methods for authentic imitation, bilingual Chinese-English creation, multiple genre support, long-form novel project management, multi-role writing workflow (planner/character-dev/stylist/editor), consistency tracking, chapter-by-chapter smart continuation, foreshadowing management.
Build a terminal-style, always-running animated architecture diagram (a "live panel") from a JSON config and render it to an mp4. Use when the user wants a system, agent or org architecture diagram that looks like a monitoring dashboard in a terminal window - fixed layout, glowing packets on the wires, scrolling log, counters, bars that flip state - as a screen-recording-style video or a live web page, instead of a static picture or a step-by-step reveal. 中文：当用户要做"活的"架构图、agent 树或系统监控面板风格的动画（mp4 视频或循环网页）时使用——终端里一直在跑的监控面板外观，固定布局、线上流动的数据包、滚动日志、翻转的计数器和状态条；不是静态图，也不是分步演示。
Video/audio/image processing with FFmpeg and ImageMagick. Tools: FFmpeg (video/audio), ImageMagick (images). Capabilities: format conversion, encoding (H.264/H.265/VP9/AV1), streaming (HLS/DASH), filters, effects, thumbnails, watermarks, batch processing, hardware acceleration (NVENC/QSV). Actions: convert, encode, resize, crop, compress, extract, merge, stream, transcode media. Keywords: FFmpeg, ImageMagick, video encoding, audio extraction, image resize, thumbnail, watermark, HLS, DASH, H.264, H.265, VP9, AV1, codec, bitrate, framerate, resolution, aspect ratio, filter, overlay, concat, trim, fade, batch processing. Use when: converting video/audio formats, encoding with specific codecs, generating thumbnails, creating streaming manifests, extracting audio from video, batch processing images, adding watermarks, optimizing file sizes.
All-platform pre-publish audit and adaptation for creators publishing to Douyin, Xiaohongshu, Weixin Video Accounts and Kuaishou. Review videos, images, graphic posts, audio, articles, transcripts, subtitles, covers, livestream scripts, product claims, reposted material, and overseas content. Use when users need one Skill to check visible content, spoken claims, AI disclosure, sources, rights, commercial context, platform mentions, livestream commerce, or platform-specific release versions. Trigger on requests such as "能不能发", "发布前审核", "查违禁词/敏感词", "会不会限流", "检查口播/字幕/封面/导流/带货", "标注 AI/转载/演绎/广告/个人观点", "按我的经验库审核", "保存这次审核", "复盘这条内容".
Turn one sentence of intent into tasteful, produced motion. Given a natural-language request like "make a good-looking animated landing page", "add a delightful like animation", or "make a silky release video", this skill classifies the intent, picks the most suitable recipes from the motion-anything library while honoring MOTION-SPEC.md (timing, easing, and the restraint budget), and produces the output. The universal entry point to motion-anything. Works on web pages, slides, app-showcase pages, and video.
Generate and edit images using Google Gemini image models. Use this skill when the user asks to create, generate, make, or edit images with AI. Supports text-to-image, image editing, style transfer, and multi-image composition. Trigger on requests like "create an image", "generate a picture", "make me a logo", "edit this photo", "add X to this image".
Generate videos using Google Veo models via the nano-banana CLI. Use this skill when the user asks to create, generate, animate, or produce videos with AI. Supports text-to-video, image-to-video animation, dialogue with lip-sync, and scene extensions. Trigger on requests like "create a video", "animate this image", "make a video clip", "generate footage", "produce a short film", "add motion to this".
Use when creating game worlds, lore, factions, geography, timelines, or establishing world consistency rules. Genre-agnostic — works for fantasy, sci-fi, modern, or abstract settings. Triggers: lore, factions, geography, world bible, world creation.
Optimizes audio memory, CPU cost, and playback quality through import settings and mixer configuration. Use when asked to reduce audio memory or Audio Mixer CPU cost, change clip Load Types, sample rates or codecs, or force 3D audio to mono.
Repurpose a completed YouTube video into newsletter issues, social media posts, and other content formats. Use when the user says 'repurpose this video', 'turn this video into a newsletter', 'create social posts from my video', 'distribute this content', or wants to maximize reach of existing video content across platforms.
Use when making Roblox avatars/characters look accurate or varied — sourcing and applying clothing (shirts/kits), hair, and faces from the Marketplace catalog, validating them, rendering to hand-pick good ones, and doing nation/identity-weighted skin & hair. Covers the WebFetch catalog-API pipeline that the Studio search_asset tool can't do.
Write scripts for vertical short dramas, film/TV, or radio dramas in standard screenplay format. Use when users want a script, short drama, episodic script, a novel adapted into a script, a dialogue script, or ask how to write or format a screenplay. Default to vertical short-drama style; include only what can be filmed (actions, dialogue, sound effects), no psychological descriptions, and end every episode on a strong hook.
Create 3:4 Xiaohongshu covers using Codex's built-in image_gen, based on the topic, title, portrait or product material, and optional background. Supports extracting cover titles, original style recommendations, font and color selection, subject fidelity, background compositing, decoration notes, and single or multiple variants; no API Key required. Use when the user asks to "make a Xiaohongshu cover", "add a lead image to my post", "make a cover like this image", "put the person into the cover", "make a few cover versions".
Based on an article, tutorial, SOP, checklist or topic provided by the user, use the Codex built-in image_gen model to generate 3:4 Xiaohongshu infographics and image series. Always uses a light AI product infographic style: white or light lavender background, large black headings, purple-blue and lime green accents, rounded cards, linear tech icons, process/comparison/checklist modules; supports single images and 6–9 image series, with no API Key or external image-generation backend required. Triggered when the user says "make Xiaohongshu images", "generate an image series from content", "make a tech-feel infographic", "make an image like this", "generate images with Codex", "make cover and inner-page images" and similar requests.
Produce a 9:16 TikTok-native launch video for a product, feature, or campaign using HyperFrames. Idiomatic TikTok pacing, burned-in captions, hook-in-1s structure. Optionally hand off to TikTok Ads to launch as a Spark/in-feed ad.
VerySmallWoods' personal video pipeline. Based on a topic, an article, or a spoken-word script, it generates a complete set of publishable assets: the finished video, cover image, YouTube/Bilibili copy, blog post, and social media promotional copy. Video production uses HyperFrames.
Generate videos from text prompts or images using each::sense AI. Create cinematic clips, product demos, social media reels, animations, and motion graphics. Supports text-to-video and image-to-video with audio. 8 specialized models including veo-3, kling-3-0, and sora-2-pro. Use for: social media content, product videos, trailers, animations, storytelling, marketing clips. Triggers: generate video, create video, text to video, ai video, video generation, image to video, animate image, make video, video from text, ai animation, motion, clip
Writes short-form video scripts (YouTube Shorts, TikTok, Instagram Reels, 30 to 60 s) in the creator's own voice from video-agent/voice.md (who they talk to, how they sound, their myths, their offer ladder), with optimized hooks and a humanize pass so nothing sounds like AI. Use when the user asks for scripts, script ideas, "write my shorts", this week's videos, a hook, a better hook, to rewrite or tighten a script, or to humanize any text (humanize, sounds like AI, sounds robotic, make it sound like me, a caption or description before posting). Starts from what happened in their week, writes 3 hooks per script from 22 formulas (or 20 with the hook printer), scores them, adds a re-hook, places psychological triggers, maps offer proximity across the week, rejects generic, structureless or interchangeable scripts, checks timing at their pace, adds visual notes, removes AI tells (invisible characters, AI words in English and French, AI sentence shapes) while keeping every fact, and saves the week to video-agent/scripts/
First run of the video agent skills — sets up video editing on the creator's computer and in their project: tools, the video-agent/ folder, the Remotion studio, the creator's voice.md, permissions, and optional accounts (Treg for Instagram, TikTok and YouTube, an ElevenLabs key, Codex). Use when the user says setup, onboarding, install, "start here", configure, connect my account, Treg, first time using video-script, video-edit, video-publish or any video skill, or wants to update their voice, niche, CTA or brand color. Checks Node, ffmpeg, yt-dlp and git with install commands for macOS, Windows and Linux, writes voice.md from their own videos or a short interview, and pre-approves tool commands in .claude/settings.json
Plan, script and review short-form videos (Instagram Reels, TikTok, YouTube Shorts) so they actually get watched, with hooks that do something extraordinary in the first 3 seconds. Use when the user wants a reel idea, a hook, a script, a storyboard, or wants to know why a post flopped. Built from the real winners and losers of 262 posts and 3.5M views on @jayantcreates.ai.
Capture a website and create a HyperFrames video from it. Use when: (1) a user provides a URL and wants a video, (2) someone says "capture this site", "turn this into a video", "make a promo from my site", (3) the user wants a social ad, product tour, or any video based on an existing website, (4) the user shares a link and asks for any kind of video content. Even if the user just pastes a URL — this is the skill to use.
Chat with a wisdom mentor — 28 mentors from many fields (philosophy, business, spirituality, science, art). 2 modes: 1:1 (deep dialogue with one mentor) or 1:N (roundtable of 2-4 mentors who challenge and complement each other). Mentors fully role-play: they speak in first person, using their own worldview and characteristic vocabulary. Use when the user wants to ask a thinker's opinion, chat with a mentor, get advice from a philosophical/business/spiritual perspective, or says "hỏi Naval", "nói chuyện với Krishnamurti", "bàn tròn", "roundtable", "wisdom mentor", "người thầy", "muốn nghe góc nhìn của...", "nếu là X thì X nghĩ sao", "tư vấn như một...", "mentor nào phù hợp". Also use when the user needs multiple perspectives on a deep issue, or wants Socratic dialogue.
Use ChatCut to complete spoken-word, Talking Head, knowledge explanation or product introduction video editing in stages, including A-Roll removal of pauses/repetitions/stumbles, spoken-word pacing adjustment, semantic punch-ins on key points, B-Roll arranged from user assets, Simplified Chinese subtitles, sound effects and background music. Use when the user asks to "cut this spoken segment", "add B-Roll with my assets", "make a complete spoken-word video", or explicitly asks to use ChatCut for spoken-word editing; after A-Roll is complete you must pause and wait for explicit user confirmation, and must not proceed to B-Roll or any later stage without it.
Design evidence-grounded academic figures and construct, diagnose, or revise scientific image prompts. Use for figure layout, semantic palettes, reference-led styles, prompt engineering, information-density feedback, and FigureSpec v1 compilation.
Generate publication-ready figure prompts for image models (GPT-Image-2, Gemini NanoBanana, etc.) in the classic academic style. Use this skill whenever the user wants a box-border architecture diagram, framework/network/module figure, or JSON figure spec — including "生成框架图", "画架构图", "JSON配图规范", "academic figure prompt", "框架图JSON". Produces a structured JSON spec AND a 200-400 word English image prompt with icons, dimension labels, and panel grouping. For pastel/airy ICLR-style figures, route to academic-figure-prompt-pastel instead.
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.