Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Three.js 3D graphics library — scene setup, geometry, materials, lighting, textures, animation, loaders, shaders, postprocessing, interaction. Use when building 3D web experiences, creating WebGL visualizations, working with GLTF models, implementing custom shaders, or adding interactive 3D elements to web applications.
Use when creating or refining 3D assets with Meshy models on Scenario via MCP: image-to-3D from one photo or 1–4 multi-view angles, text-to-3D, retexturing an existing GLB, remeshing to a target polycount, UV unwrapping, auto-rigging a humanoid character, or applying a library animation clip. Keywords: Meshy 7 and Meshy 6, Ultra mode, Smart Topology, PBR maps, texture prompt, triangle or quad topology, game-ready mesh, A-pose, T-pose, GLB pipeline
Expert patterns for Godot audio including AudioStreamPlayer variants (2D positional, 3D spatial), AudioBus mixing architecture, dynamic effects (reverb, EQ, compression), audio pooling for performance, music transitions (crossfade, bpm-sync), and procedural audio generation. Use for music systems, sound effects, spatial audio, or audio-reactive gameplay. Trigger keywords: AudioStreamPlayer, AudioStreamPlayer2D, AudioStreamPlayer3D, AudioBus, AudioServer, AudioEffect, music_crossfade, audio_pool, positional_audio, reverb, bus_volume
Repair imported ComfyUI workflows with missing or broken model metadata for ComfyUI-RunpodDirect. Accept a workflow JSON or original ComfyUI output PNG, extract and inventory it, and resolve exact model artifacts. Use when a user reports a broken workflow, missing models, or failed model discovery; use companion-clis for generic Hugging Face transfers and runpod-mcp or runpodctl for Pod lifecycle.
Edits a talking-head take end to end into a paper-style motion design short (1080x1920, also 1:1 and 16:9) with its own render engine (Node + Chrome + ffmpeg, free, offline). Use when the user says edit, cut, captions, subtitles, motion design, montage, "make it a short", render, or gives a recording from video-agent/takes/. Transcribes locally with word timings (whisper.cpp, or ElevenLabs when a key is set), cuts every silence, lays the boards on a textured paper above the camera card (counters, terminals, logos, trees, real screens, HTML motion components), word-by-word captions, a calm music bed, a few designed hits, -14 LUFS, and writes the mp4, an .srt and the post copy.
Build an ultra-realistic open ocean in Three.js with a deep-water Gerstner spectrum shaded per pixel from analytic derivatives, each wave faded at its own pixel footprint, plus Fresnel sky-cube and planar reflections, subsurface light through the crests, HDR sun glitter, Jacobian lace foam, a Kelvin wake, hull foam, buoyancy queries and aerial perspective into the sky's horizon haze. Use for seas, oceans, sunset or daylight water, sun glitter paths, ship or boat scenes, wakes, coastal flyovers and any 3D scene with a moving camera and a water horizon. Includes the approved water from Pirate Ship Sunset as a reusable module and demo.
Create educational videos using Remotion with Kurzgesagt/回形针 style. Use when users want to: (1) create teaching or educational videos, (2) design video storyboards, (3) produce animated explainer videos, (4) build SVG-based animations for learning content, (5) visualize complex concepts with motion graphics, (6) make science/tech explainer videos, (7) create 可视化讲解视频 or 科普视频. Triggers on requests like '帮我做一个教学视频', 'create an explainer video about X', '制作科普动画', 'make a video explaining Y'. This skill requires remotion-best-practices skill for technical implementation.
Expert blueprint for procedural content generation (dungeons, terrain, loot, levels) using FastNoiseLite, random walks, BSP trees, Wave Function Collapse, and seeded randomization. Use when creating roguelikes, sandbox games, or dynamic content. Keywords: procedural, generation, FastNoiseLite, Perlin noise, BSP, drunkard walk, Wave Function Collapse, seeding
Generate logo variants for a brand, producing multiple production-grade marks across different architectures (wordmark, lockup, symbol-only, letterform-as-symbol, monogram) with rationale, application specs, and decision-ready presentation. Distinct from `brand-identity`, which produces the complete system (logo plus color, type, voice, applications). Goes deep on the logo specifically: typographic register, symbol approach, application-context discipline (16px favicon, embroidery, signage, motion), and production specs (SVG-ready, single-color, reverse). Triggers on logo design, design a logo, wordmark, monogram, lockup, mark variants, logo iterations, redesign or refine the logo, logo refresh, letterform-as-symbol, and exploring mark variants for an established brand. Does NOT fire for full brand identity (use `brand-identity`), positioning or strategy (use `brand-discovery`), project-wide creative direction (use `creative-direction`), or one-off illustration
Turn a full script, story, or voiceover draft into structured, cinematic video-generation prompts — automatically segmented scene-by-scene. Best for structured-prompt video tools like Seedance 2.0, Kling, and Jimeng (即梦). Use this skill whenever a user pastes a script / story / 口播稿 / 分镜脚本 and wants it turned into video prompts, asks to write a video prompt, generate a storyboard, split a script into shots, or describe a cinematic scene for video AI. Each segment gets a complete prompt: 基础设定 (characters), 道具 (props), 场景 (scene), 声音 (sound), 氛围与画质 (mood & quality), 画面内容 (shots — auto-decomposed into a shot list or a long take, each with 景别/构图/运镜). Also trigger on keywords: video prompt, video generation, AI video, text-to-video, storyboard, 分镜, 拆分镜, 视频提示词, 视频生成, 脚本转视频, Seedance, Kling, 即梦, cinematic prompt, shot list, camera movement. Trigger even if the user only has a rough one-sentence idea — treat it as a single segment and flesh it out.
Use when a creative task has no direction yet: finding inspiration, exploring looks or styles, proposing concept directions to pick between, building a moodboard or reference board, doing visual research, mining existing assets for a style, or escaping generic AI-looking output. Also when a saved Scenario collection should drive a new batch as art direction. Keywords: inspiration, moodboard, references, art direction, style exploration, concepts, serendipity, ideation, surprise me.
REQUIRED for all image generation requests. Generate and edit images using Nano Banana (Gemini CLI). Handles blog featured images, YouTube thumbnails, icons, diagrams, patterns, illustrations, photos, visual assets, graphics, artwork, pictures. Use this skill whenever the user asks to create, generate, make, draw, design, or edit any image or visual content.
Use Gemini Omni Flash Preview through AnyCap for practical short-video editing workflows. Trigger when a user wants to edit or refine an existing video with natural-language instructions, especially product replacement, object removal or replacement, relighting, restyling, preserving a person/scene while changing one element, or using reference images with a source video. Covers model/schema discovery, prompt construction, AnyCap CLI command patterns, scenario-specific templates, video QA, and retry strategy for `gemini-omni-flash-preview` / `edit-video`
Make or edit animated, narrated, interactive web books from reference PDFs, animated picture books (paper-puppet films) from a story, a joke or a fairy tale, narrated documentaries about a book or an idea, built from public-domain pictures, and upright shorts for a phone feed that make one point in under three minutes. Use for book planning, storyboards, chapter animation and exercises, narration, and cover and contents pages, including "make the next chapter", "turn this story into a picture book", "make a documentary about this book", "make a short about this for YouTube Shorts" or "fix this animation". Deliver runnable static book files; bookshelf design, bookshelf registration and hosting are separate tasks.
The scriptwriter of the creator's videos - short-form scripts (YouTube Shorts, TikTok, Instagram Reels, 30 to 60 s) in their own voice from video-agent/voice.md, each on a concept proven by the niche's outliers, with 3 scored hooks from 22 formulas, a re-hook, triggers, offer proximity, quality gates, timing at their pace and visual notes, then a humanize pass (invisible characters, AI words in English and French, AI sentence shapes, facts kept). Modes - the week (default), one topic, a funnel or divisive script, a making-of, hook printer, rewrite, humanize, and viral (what over-performs in their niche, scored against each account's median, YouTube free with yt-dlp, TikTok and Instagram through Treg, into video-agent/swipe.md). Use when the user asks for scripts, ideas, this week's videos, a funnel or divisive (clivant) script, a making-of, a hook, a better hook, to rewrite a script, what goes viral, viral hooks, what works in my niche, competitor research, swipe file, outliers, or to humanize any text (sounds like AI, make it sound like me, a caption before posting).
Images as the deliverable on Moda: generate, edit, or transform — text-to-image, reference-guided, generative edit, upscale a photo, outpaint/uncrop, reframe a photo/image. Use for: “generate an image,” “remove the background,” “upscale this photo.” NOT: imagery inside a design you're building — that format skill generates its own; animating an image → moda-video; logo design → moda-social. Metered.
Senior graphics programmer specializing in Three.js, WebGL, shaders, and performance optimization. Use when "three.js, threejs, webgl, 3d web, 3d graphics, shaders, glsl, 3d scene, 3d animation, threejs, webgl, 3d-graphics, shaders, animation, interactive, web-graphics" are mentioned.
Deepgram API reference for speech-to-text, text-to-speech, voice agents, audio intelligence, and account management. Use whenever building with Deepgram APIs — REST or WebSocket. Covers authentication, all endpoints, query parameters, request/response schemas, and WebSocket message formats. Reference files are organized by domain: listen (STT — Nova and Flux STT), speak (TTS — Aura and Flux TTS), agent (voice agents), read (text/audio intelligence), models, projects, auth, and self-hosted
Use-case-driven multi-step pipelines on fal.ai. Trigger when the user asks for a specific kind of content production rather than a single endpoint call: "make a commercial", "ad creative", "product photography", "cinematic shot", "film look", "character design", "consistent character", "anchor system", "storyboard", "multi-shot", "narrative video", "talking head", "lip sync", "make this person talk", "virtual try-on", "garment transfer", "restore image", "deblur", "denoise", "fix face", "old photo restore", "add audio to video", "video sound effects", "product shot", "photoreal", "realistic photo", "candid photo", "editorial portrait", "documentary photo", "looks like a real photograph", "iPhone-style photo", "film photo", "archival photo". Each recipe describes inputs, the genmedia call sequence, and quality checks.
Generate audio on Moda: voiceover/TTS, narration, music, jingles, sound effects — up to 10 minutes per render, delivered as a file (no canvas slot). Use for: voiceover, narration, "read this aloud", jingle, background music, SFX. Pairs with moda-video-clip for scored video. Metered.
Plan and orchestrate end-to-end video production pipelines in ComfyUI with validation gates and error recovery. Handles img2vid, txt2vid, vid2vid, and multi-shot video production. Produces pipeline plans with correct step ordering (generate, validate, animate, validate, concat), model selection, retry strategies (seed randomization, parameter adjustment, model fallback), and VRAM-aware resource management. Use when asked to make a video, animate images, create a multi-shot video, set up a video pipeline, or orchestrate video production in ComfyUI. Does NOT cover still image generation, prompt writing, workflow building for non-video tasks, video editing in external tools, model training, installation, or hardware recommendations.
Use when lighting, rendering or compositing in Blender: light a character, product or hero shot, interior at dusk or night, three-point or motivated lighting, sun and sky, HDRI, EEVEE vs Cycles settings, light leaks or noise, flat or muddy renders, subject not standing out, AgX/ACES color management, render passes, EXR, light groups, bloom and vignette in the 5.x compositor, rule of thirds and value checks, toon/NPR looks, or judging a render like a lighter.
Generate whiteboard animation videos from images. Convert any color image into an animation with line drawing and coloring stages, featuring a hand overlay, and output an H.264 MP4 video. Supports single-image and batch modes. Trigger when the user says “turn an image into a whiteboard animation,” “whiteboard animation,” or “batch whiteboard animation”
Use when generating 3D assets with Rodin Hyper3D models on Scenario via MCP: image-to-3D from up to five multiview stills, text-to-3D from a prompt, fast prototyping versus full quality tiers, quad or triangle topology, PBR materials, HighPack 4K textures, T-pose or A-pose characters for rigging, or splitting a finished mesh into parts and retexturing it with Bang!. Keywords: Rodin Gen-2.5, Hyper3D, Deemos, image to 3D, text to 3D, part segmentation, texture delight.
An agent creates an animated film of an average session with its user, from beginning to end, using the user's real messages—whatever their routine involves, such as coding, writing, data, design, or video. The agent sits at a desk in a room decorated based on what it knows about the user, while the monitor shows the user's kind of work. Requests glide in as paper planes and unfold. Corrections appear as a 3D cartoon mallet that bonks, then turns to reveal words on its face. Praise is a butterfly towing words on a ribbon. The agent types each reply onto a sheet, folds it, and throws it back. The film opens with the agent asleep until the monitor boots: “Initiating session with <name>”. The agent composes its own orchestral score, cued to every action. Made with p5.brush watercolour and rendered in HyperFrames. Trigger on: “a video of our session,” “animate a session with me,” “session story,” “show what working with me is like,” “make a film about me and my agent.”
Sets up the creator's video editing studio, for shorts and YouTube videos, on their computer and in their project - tools, the video-agent/ folder, the render engine, the creator's voice.md, their default editing style (an onboarding on their colors, paper, fonts, captions, sound, Instagram CTA, real tool logos, logo and end card, with a preview on their own footage), permissions, and the 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 studio-script, short-editing, publish or any video skill, or wants to update their voice, niche, CTA, brand color, look, font, captions, sound effects, logo or tool logos (`/studio-setup style`). 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, pre-approves the tool commands in .claude/settings.json.
Generates images with Codex (OpenAI's coding agent and its built-in image tool, included in a ChatGPT plan) — YouTube thumbnails, covers, B-roll stills, board images, profile pictures — one Codex job per image, several in parallel, with the creator's photos as reference when the face matters, then copies them into the project with a manifest. Use when the user says generate an image, a thumbnail, a cover, make me a visual, codex images, génère une image, miniature, or when publish or short-editing needs an image.
A generator for line-by-line annotations of classical Chinese, combined layouts, chapter theme diagrams, and full-chapter interpretations. Arranges the source text, word and phrase annotations, sentence explanations, an unlettered illustration at the top, and the chapter interpretation into one readable, long PNG. Use when the user invokes ljg-classic or requests line-by-line annotations, color-coded interlinear notes, chapter interpretations, illustrated classical Chinese handouts, or chapter illustrations for classical Chinese prose, poetry, or the Four Books and Five Classics. Not for translating a single line of classical Chinese, writing only modern prose, or creating generic content cards.
Generate publication-quality academic illustrations through a local Codex app-server bridge that uses Codex native image generation. This is a separate experimental alternative to `paper-illustration`, intended for Claude Code users who want a GPT-image-style renderer without modifying the original skill.
Generate professional product packaging designs using each::sense AI. Create box designs, food packaging, cosmetic containers, beverage labels, supplement bottles, coffee bags, candle packaging, gift boxes, shopping bags, and product labels.
Use when generating navigable 3D worlds or Gaussian splat scenes on Scenario via MCP: text-to-world, image-to-world, a 360 panorama or skybox turned into a walkable splat, multi-view photos or a video walkthrough reconstructed as a splat, or one object photo to splat. Keywords: World Labs Marble 1.1, 1.1 Plus, 1.0 Draft, Hunyuan World, HY World, image-to-skybox, skybox-to-splat, multiview-to-splat, TripoSplat, gaussian splatting, 3DGS, spz, world generation, explorable scene, VR.
Expert blueprint for TileMapLayer and TileSet systems for efficient 2D level design. Covers terrain autotiling, physics layers, custom data, navigation integration, and runtime manipulation. Use when building grid-based levels OR implementing destructible tiles. Keywords TileMapLayer, TileSet, terrain, autotiling, atlas, physics layer, custom data.
Extract and document a writer's distinctive voice patterns for consistent reproduction. Use when you need to capture writing voice, analyze writing style, create a voice guide, or write in someone's established style. Keywords: voice, tone, style, writing analysis, fingerprint.
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.