Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Add particle effects (smoke, fire, sparks, magic, glow, dust, snow, drips, bubbles, bursts) to Blockbench models with MCP tools. Covers Bedrock/Snowstorm particle JSON, presets and design knobs, locators, particle keyframes on animations, Blockbench's Animate-mode preview, and delivery to Bedrock resource packs (client entity particle_effects) and GeckoLib, plus fallbacks for glTF, Java block and Hytale targets that cannot carry particles. Use when a model needs an emitter, VFX, particle effect, locator, Snowstorm file, particle_effects keyframe, or effect polish in a deliverable.
Analyze character traits based on story text and generate detailed character profiles. Use to understand character settings, provide actors with references for shaping roles, and create character dossiers for scriptwriting. Use when the user mentions character profiles, character dossiers, or character analysis.
Motion and creative-coding skill package covering Three.js, GSAP ScrollTrigger, React Three Fiber, Framer Motion, Babylon.js, A-Frame, PixiJS, Locomotive Scroll, Barba.js, React Spring, Magic UI, AOS, Anime.js, Lottie, Rive, Spline, and a Blender-to-web pipeline. One of the broadest motion/creative suites available.
Bridges asset requirements from motion design specs to production-ready assets. Parses specs for required assets, recommends free/paid sources, provides format conversion guidance, generates validated import code, and offers asset preparation checklists. Use when preparing assets for Remotion projects or when asked "where to get assets", "how to prepare assets", "asset formats for Remotion".
World-class character rigging and animation systems expertise: skeleton hierarchies, deformation, FK/IK, facial rigs, weight painting, and game engine integration from someone who has shipped AAA characters. Use when terms such as “rig a character”, “character rigging”, “skeleton hierarchy”, “weight painting”, “skin weights”, “joint orientation”, “fk ik”, “inverse kinematics”, “forward kinematics”, “facial rig”, “blend shapes”, “morph targets”, “twist bones”, “deformation”, “animation retargeting”, “root motion”, “bone limits”, “control rig”, “humanoid avatar”, “generic rig”, “candy wrapper effect”, “volume preservation”, “rigging”, “animation”, “character”, “skeleton”, “deformation”, “maya”, “blender”, “unity”, “unreal”, “fbx”, “skinning”, “weight-painting”, “ik”, “fk”, “facial”, or “mocap” are mentioned
Use when animating characters or scenes in Godot 4.7: AnimationPlayer clips and RESET, AnimationTree state machines and blend spaces built in code, Mixamo or glTF import, loop settings, BoneMap retargeting and shared animation libraries, root motion, foot sliding, foot IK with TwoBoneIK3D or custom SkeletonModifier3D, tweens, SpringArm3D cameras, cutscenes; or when a character T-poses, slides, floats, freezes mid-blend or a transition never fires.
Give the project one visual point of view, then generate images against it. Use before generating, editing or replacing any image asset — hero image, og:image, illustration, empty state, avatar, background, card art, favicon source. Also use when generated images look generic, "AI", off-brand, or when a set of images do not look like they belong to the same product. Skip for UI layout, CSS and component work, which belong to frontend-design and impeccable.
Analyzes article structure, identifies positions requiring visual aids, generates illustrations with a Type × Style two-dimension approach. Use when the user asks to "illustrate article", "add images", "generate images for article", or "为文章配图".
Generates article cover images with 5 dimensions (type, palette, rendering, text, mood) combining 9 color palettes and 6 rendering styles. Supports cinematic (2.35:1), widescreen (16:9), and square (1:1) aspects. Use when user asks to "generate cover image", "create article cover", or "make cover".
Use when asked to create an explainer, walkthrough, or board-work video, worked-example lesson, or mini-lesson for a math problem (including geometry, algebra, functions, and motion/行程 problems) from a screenshot or text, with Chinese voice-over. Deliver a browser preview (HTML) first; export MP4 only if requested. Choose from five visual templates: hand-drawn canvas animation with bilingual zh+en subtitles, or one of four board styles (chalkboard, geometry lab, whiteboard, or graph-paper notebook, with auto-rotating 3D figures). Choose a TTS engine: 智谱 GLM-TTS, 豆包 Doubao, MiniMax, or free edge-tts / macOS say
Use whenever the user asks to "humanize", "make this sound more human", "rewrite to avoid AI detection", "make this less AI-sounding", "add a human voice", or "write like a person". Also use when the user pastes text and asks why it reads as robotic, generic, flat, or AI-like, or when generating new text in a register where AI tells (em dashes, semicolons, hedges, banned vocabulary like "delve", "leverage", "robust") would damage credibility.
Create, inspect, recolor, and repair Lottie animations, including converting SVG art into animated Lottie JSON. Use when asked to animate a logo, icon, or SVG; build motion graphics, micro-interactions, loaders, spinners, success/error states, toggles, or entrance animations; produce or edit a .json/.lottie animation file; re-skin one in a brand palette or dark mode; add wiggle, bounce, pulse, fade, scale, rotate, morph, or trim-path draw-on effects; build walk cycles or character rigs; or debug a Lottie that renders blank, jumps at the loop, or looks wrong in a player.
Perform a multidimensional preliminary assessment based solely on the story text provided by the user (market potential, originality, and content strengths). Use for preliminary screening of novels or stories and quality assessment without external information. Do not use when online research into popularity, author background, or rights data is needed; use ip-evaluator instead.
Download source video or audio from Douyin, YouTube, Bilibili, Twitter/X, Xiaohongshu, and other yt-dlp-supported URLs into the content-creation workspace. Use when the user says 下载视频, 下载音频, 保存这个链接, 把源视频拉下来, or when another video workflow needs a local source file before transcription, frame analysis, editing, reference replication, or archiving.
Use when rewriting AI-generated text to match a specific person's or brand's authentic voice. Use when AI output sounds generic, corporate, or detectable. Unlike generic "humanizers," this skill requires voice analysis input and produces voice-consistent output, not fake imperfections.
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.
Prose editing, rewriting, and humanizing text for natural tone, or auditing a draft for AI tells without rewriting. Use when asked to write, rewrite, edit, humanize, proofread, fix tone, remove AI language, or check whether writing reads as AI. For copy, docs, blog posts, emails, or PRs.
Skills for AI media generation through the muapi.ai platform: cinematic video direction, image generation with Nano Banana, Seedance 2.0 video, UGC and product video ads, logos, UI mockups, YouTube thumbnails and shorts, social posts and AI clipping. Routes to models such as Midjourney v7, Flux Kontext, Kling 3.0 and Veo3. Requires a muapi account and the muapi CLI.
Generate Chinese illustrations in Ian’s style. Use for requests to create illustrations for Chinese articles, posts, blogs, Notion documents, workflow documents, methodologies, processes, structures, states, metaphors, or viewpoints, including requests for “怪诞,” “小黑,” “手绘,” “正文配图,” “文章插图,” “配图建议,” “shot list,” or “去标题/改图.” By default, use the 小黑 IP, pure-white hand-drawn style, a few red, orange, and blue annotations, and a clean, fresh yet imaginative visual style.
Logo design principles and AI image generation best practices for creating logos. Covers logo types, prompting techniques, scalability rules, and iteration workflows. Use for: brand identity, startup logos, app icons, favicons, logo concepts. Triggers: logo design, create logo, brand logo, logo generation, ai logo, logo maker, icon design, brand mark, logo concept, startup logo, app icon logo
Open-source, local-first alternative to Claude Design. A daemon plus web UI or desktop app that runs design skills (web prototypes, SaaS landing pages, dashboards, mobile onboarding, decks, email, social carousels, posters, motion frames, sprite animation) and brand-grade design systems on top of your own coding-agent CLI or API key.
End-to-end Chinese video rewriting pipeline. Use when the user provides a Bilibili, Douyin, Xiaohongshu, YouTube, or web video URL, or a local video file, and asks for video script rewriting, video remixes, rewriting and producing a video, rewriting a linked video, extracting a verbatim transcript and rewriting it, or producing a new AI video from a source video. Orchestrates ra-逐字稿提取skill and ra-洗稿, ending at the 01-内容生产/视频工作台/待制作/ handoff queue; production continues through ra-video-production-director only on explicit request (直接制作/直接出片/一条龙).
Use when building 3D levels and worlds in Godot 4.7: third-person or FPS controller, player stuck on stairs, spring arm camera clipping, blockout with CSG, GridMap and MeshLibrary, terrain (heightmap, HeightMapShape3D, Terrain3D regions, textures, painting), scattering trees with MultiMesh, streaming a big world, procedural meshes with SurfaceTool, Jolt physics settings, importing GLB or glTF from Blender or AI generators (Scenario, Tripo, Meshy, Hitem3D), import scripts, name suffixes, model facing, collision generation, LODs.
Generate cloned narration for the content workspace. Use local IndexTTS2 with the canonical lossless Pluviobyte reference for every local video, voice audition, digital-human narration, and subtitle-ready audio render. Preserve MiniMax only for public-facing articles or tutorials that teach users to connect through the relay service; never use MiniMax for local video production.
Use when the user has one or more video clips and wants to add post-production on top — AI-generated cover as first frame, HTML/CSS captions synced to SRT, kinetic illustration overlays at hook moments, chapter chips, end-card CTA, or any other timed motion graphics. Most often used as the downstream of `/wjs-segmenting-video` — pick up where that skill stopped (raw cropped clip + per-clip SRT) and produce the upload-ready MP4. Backed by HyperFrames so everything compiles to ONE final encode — no cascade of re-encodes. Triggers — "加封面", "加字幕", "加动画", "加 CTA", "做后期", "post-production", "title card", "kinetic captions", "end card".
Use when the user wants to teach or learn an English word as a video — turn a single English word into a self-contained HLS “supercut” lesson built from the mira video base. Stitches every season2 clip where the word is spoken (via the search-app API) into one .m3u8, prepended with a Claude-written bilingual word-intro card (word + IPA + Chinese gloss + usage, Volcano TTS) and appended with a 关注王建硕 CTA card. No MP4 burn. Triggers — “teach <word>”, “讲讲 <word>”, “学英语 <word>”, “把 <word> 做成视频”, “/wjs-teaching-english <word>”
Generate AI music and songs with ElevenLabs, Diffrythm, Tencent Song Generation via inference.sh CLI. Models: ElevenLabs Music (up to 10 min, commercial license), Diffrythm (fast song generation), Tencent Song Generation (full songs with vocals). Capabilities: text-to-music, song generation, instrumental, lyrics to song, soundtrack creation. Use for: background music, social media content, game soundtracks, podcasts, royalty-free music. Triggers: music generation, ai music, generate song, ai composer, text to music, song generator, create music with ai, suno alternative, udio alternative, ai song, ai soundtrack, generate soundtrack, ai jingle, music ai, beat generator, elevenlabs music, eleven labs music
Bake the edited voiceover into a finished video file: burn the ledger segments, zoom-ins, subtitles, and HTML visual layers into the mp4 all at once. Use when the user says export, produce the final cut, burn in subtitles, render, export video, or generate the final file. Do not use for generating word-deletion candidates, writing subtitles, or creating visual animations.
Add visuals to the edited voiceover: overlay HTML layers on the screen recording (highlight annotations / full-screen small black animations / zoom-ins), bind layers to the subtitle screen, drive them frame by frame from the player, and view them directly in the preview. Use when users ask for a storyboard, visuals, animations, highlighted callouts, B-roll, or a storyboard. Do not use for word-deletion editing, subtitles, physical cuts, or final video rendering.
Granular feedback on drafts without rewriting. Generates highlighted HTML with click-to-reveal inline comments. Use when user says "comment on this", "leave comments on", "give feedback on", or asks for feedback on a draft. Supports multiple lenses—editor feedback, POV simulation ("as brian would react"), or focused angles ("word choice only", "weak arguments"). A granular alternative to rewrites that lets users review feedback incrementally without losing their voice.
Explainer video production guide: scripting, voiceover, visuals, and assembly. Covers script formulas, pacing rules, scene planning, and multi-tool pipelines. Use for: product demos, how-it-works videos, onboarding videos, social explainers. Triggers: explainer video, how to make explainer, product video, demo video, video production, video script, animated explainer, product demo video, tutorial video, onboarding video, walkthrough video, video pipeline
Generate videos with Google Veo models via inference.sh CLI. Models: Veo 3.1, Veo 3.1 Fast, Veo 3, Veo 3 Fast, Veo 2. Capabilities: text-to-video, cinematic output, high quality video generation. Triggers: veo, google veo, veo 3, veo 2, veo 3.1, vertex ai video, google video generation, google video ai, veo model, veo video
Use when the user wants copy-ready AI scientific illustration prompts based on papers, research documents or text, figure captions, proposal or defense materials, or reference images that can be interpreted visually. Triggers include paper figures, scientific illustration, graphical abstracts, mechanism diagrams, technical roadmaps, experimental setup diagrams, model architecture diagrams, multi-panel comparisons, presentation overview figures, reference-image style transfer, creating templates for unfamiliar fields, or uploading or citing PDFs, Word, LaTeX, Markdown, abstracts, methods sections, figure captions, project descriptions, research plans, or reference images. Also use for scientific figure or research illustration prompts, graphical abstract prompts, mechanism diagrams, technical roadmaps, experimental setup diagrams, model architecture diagrams, multi-panel scientific figures, paper or conference figures, or AI image prompts based on papers, research documents, abstracts, methods sections, figure captions, proposals, reference images, PDFs, Word docs, LaTeX files, Markdown files, or research plans.
Production pipeline for interactive and generative visual art using p5.js. Covers 2D/3D rendering, noise and particle systems, flow fields, GLSL shaders, pixel manipulation, kinetic typography, WebGL, and audio-reactive visuals, with headless export to HTML, PNG, GIF, MP4 or SVG. From NousResearch's Hermes agent.
Create or revise Chinese social media covers using the registered Skill-封面 asset system, reusable styles, presenter identities, ratio-specific layouts, and precise replacement prompts. Use only when the user explicitly says “skill-cover”, “Skill-封面”, or “实战风格”; requests registered style or gesture assets; asks for simultaneous 3:4 and 4:3 registered-layout output; replaces a reference cover’s person with the current digital-human frame; or extends this registered cover system. Do not use for generic “制作封面/生成封面” requests; use dog-skill for generic cover requests.
Static visual content across 20+ formats — diagrams, mermaid, infographics, D3 dashboards, comics, icons, wallpaper — via Nano Banana Pro (default), Nano Banana, and Flux. Use when art, illustration, diagram, flowchart, infographic, header image, blog social thumbnail, visualize, generate image, mermaid, architecture diagram, comic, icon, blog art, framework diagram, D3 chart, remove background, wallpaper. Not for locked house-style YouTube/channel/video thumbnails, video or animation (use Remotion), or web UI design and integrated frontend layout (use Webdesign).
Generates self-contained HTML/CSS animations of app features for walkthroughs, demos, and onboarding. Researches the target app via Chrome, interviews the user, generates a structured brief and animation HTML file, then enters an iterative freeze-inspect-feedback review loop until the user approves. Use when the user says “css animation”, “animate this feature”, “create a css walkthrough”, “animation walkthrough”, or wants to create a CSS-based visual demo of an app feature
Create dark-themed, animated technical diagrams as self-contained HTML+SVG files—flowcharts with visibly flowing connectors and architecture diagrams where requests travel through the system as light dots (Diagrid/Temporal landing-page style). Use whenever the user asks for a flowchart, workflow, pipeline, process diagram, state machine, system architecture, infrastructure, cloud, microservices, or network topology diagram—especially when they mention “animated,” “flowing,” “dynamic,” “alive,” “GIF-like,” or want a diagram for a landing page, README, docs, or product demo. Also use it to convert Mermaid source (a mermaid code block or .mmd file) into an animated diagram: “animate this mermaid,” “make this flowchart move.” Prefer this over static diagrams whenever the diagram represents anything that moves: requests, events, data, jobs, messages, or control flow.
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.