Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Use when one image must be composed from several reference pictures through the Scenario MCP server, each lending one thing: the character from one, the place, pose, style, light, palette, material, outfit, or layout from others. Also when a reference leaks into the result (its object, colors, or face showing where it should not) or a reference's role is ignored. Keywords: multi-reference, reference roles, combine references, style reference, pose reference, give each picture a job.
Base-layer skill for the SenseNova-Skills project, providing low-level APIs for image generation, recognition (VLM), and text optimization (LLM). This skill does not preprocess inputs; it only calls backend services and returns results. This skill is not user-facing and is intended for upper-layer skills only.
TypeGPU and raw WebGPU adapter patterns for HyperFrames. Use when creating GPU-rendered compositions with TypeGPU, raw WebGPU, WGSL fragment shaders, compute pipelines, liquid glass effects, particle systems, or any canvas layer driven by navigator.gpu that responds to HyperFrames hf-seek events.
Convert text to speech (TTS). Uses the Volcano Engine Doubao speech synthesis API and supports streaming synthesis, multiple voices, speech rate/pitch/volume adjustment, Markdown filtering, and LaTeX formula narration. Use this skill when the user wants to convert text to speech, generate read-aloud audio, create voice-overs or narration, make announcements, or mentions “text to speech”, “TTS”, “speech synthesis”, “read aloud”, or “voice-over”.
Backend-neutral image generation: create new raster images and looping GIF/WebP animations through the local one-file image-use CLI (formerly chatgpt-imagegen), using the user's ChatGPT subscription by default, the Codex backend as fallback, or an optional Gemini subscription — no API key or daemon. Triggers: image generation, generate an image, draw a picture, 画图, 画一张, 生成图片, 生图, 配图. Use for photos, illustrations, icons, hero banners, mockups, sprites, concept art, animation loops, and figures for documents, proposals, blog posts, or READMEs; save outputs in the workspace. Auto mode uses the logged-in ChatGPT browser via chrome-use. Gemini users can pass --backend gemini or agy. Proactively propose useful figures while authoring long-form content. Do not use for editing existing images, SVG/vector work, code-native graphics, established icon systems, explicit high-quality or transparent API output, or end-user image-generation services.
Use when animating a character or prop in Maya through Python: a walk or run cycle, a jump or landing, a heavy lift, throw or punch, an acting or dialogue shot, lip sync, eye darts and blinks, stepped blocking, splining, polish, Graph Editor tangents, animation layers, mocap cleanup or retargeting, game clips and root motion, or reviewing a playblast. Also when motion looks floaty, weightless, mushy, poppy, posey, sliding or chattery, or feet slap or skate.
Use when skinning a character in Maya ("bind skin", "fix the weights", weight painting, Skin Tools or ngSkinTools layers), when deformation looks wrong (pinching, candy-wrapper wrist, collapsing elbow or shoulder, volume loss, clothes poking through), when mirroring, copying or capping skin weights for Unreal or Unity, building blend shapes, corrective shapes or pose space deformation (Pose Editor on IK-driven joints), a face rig with shapes or joints, multiple skinClusters, Delta Mush or the ML Deformer, or running a range-of-motion deformation test.
Use when shading an asset in Maya with Arnold: OpenPBR or aiStandardSurface materials, hooking up Substance Painter or UDIM texture sets, color spaces (sRGB, Raw, ACEScg), a normal map that looks wrong, displacement with black holes or faceting, skin SSS, chrome and metals, glass or crystal, car paint, aiStandardHair, porting Standard Surface to OpenPBR, or a lookdev turntable with chrome and gray balls. Also when textures render washed out, too dark or plastic in Arnold.
Use when modeling in Autodesk Maya, such as hard-surface props, weapons, vehicles, a sci-fi crate, a game-ready or SubD film model, or a character or creature base mesh, or when asked to "clean up this model" or a scene before rigging or UVs. Also for support loops vs bevels vs creases, Smart Bevel, weighted normals, hard edges, pinching or wobbly highlights, a faceted cage, flipped faces, a triangle budget, n-gons, non-manifold geometry, mirroring a half, a low poly from a high poly, or deformation-ready topology.
Provider-independent title design and kinetic typography direction for video, animation, ads, explainers, social hooks, brand films, lyric/text videos, title cards, main titles, lower thirds, captions-as-design, and other motion-led text. Use when an agent must plan, prompt, implement, hand off, iterate, or QA typography hierarchy, readability, timing, reveal/hold/exit, motion semantics, line breaks, safe areas, localization, brand constraints, accessibility, flashing/motion safety, or production specs for animated text.
A content creation tool for social media creators that helps write high-quality articles and headlines, and provides guidance for generating accompanying and cover images. Supports the styles of WeChat Official Accounts, Xiaohongshu, and Douyin. Creates in depth based on 11 content insight dimensions, including core ideas, persuasion strategies, emotional triggers, memorable quotes, emotional arcs, and engagement hooks. Activates when users request content creation, such as writing an article, creating a WeChat Official Account article, a Xiaohongshu post, Douyin copy, or coming up with a headline, or mention social media content or viral copy. Also applies when users provide a topic or source material and ask for complete content.
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
Use this skill when OpenStoryline is already installed and the user wants to start the local MCP/Web services, create or continue a session, send editing instructions, perform multi-turn re-editing, and verify rendered video outputs. Also use for Chinese requests like “启动 OpenStoryline”, “把 OpenStoryline 跑起来”, and “用 OpenStoryline 剪视频”.
Plan your Lens for Learn AR with SPECS, the Devpost learning hackathon. A short interview checks your idea against AR design best practices and what Lens Studio's preview can do, then writes devpost/design-brief.md for CLAD to build from. Run it from inside your SPECS Lens Studio project folder.
Short-video hook (opening) type selection and writing, Chinese name “钩子” (hook). Use when coming up with or writing hooks, when stuck on hook ideas, designing a hook for the first 5 seconds, writing a video or spoken-script opening, or when the user asks how to make a video opening, wants opening templates or hook types, or asks for a hook. Also use during ra-洗稿 script rewriting and topic development. Provides a scenario-based taxonomy of 7 hook types with a decision flow, prerequisites, sentence templates, and common mistakes. To diagnose and optimize an existing hook draft, use dbs-hook instead; this skill selects the type, while dbs-hook polishes its execution.
Use when lighting or rendering in Maya with Arnold: a product hero shot, a character close-up or a sequence (master and shot lighting), key, fill and rim placement, "the Arnold render is noisy" or fireflies, render settings and time budgets, AOVs, light groups and EXRs for Nuke, OIDN or noice denoising, ACES or OCIO color, renders that look flat or too dark, Render Setup layers, batch or command-line renders, or judging a lit frame like a lighting supervisor.
Plan, produce, adapt, finish, deliver, and improve platform-ready short vertical social videos from a brief, script, transcript, or source media. Use for TikTok videos, Instagram or Facebook Reels, YouTube Shorts, LinkedIn vertical clips, and cross-platform social cutdowns when the work requires audience and objective definition, an honest hook and promise, short-form beat or shot planning, UI-safe layout, captions and accessibility, speech/music/SFX mixing, platform-specific exports, rights and AI or sponsorship disclosure, analytics-led retention iteration, or final QA and repair. Do not use as a provider-specific image, video, voice, or music generation guide.
Use when the user has 2+ recordings of the same event (each with a `.sync.json` sidecar from wjs-syncing-multicam) and wants them combined into a single MP4 — auto-switching between cams second-by-second on audio energy, with optional picture-in-picture inset. Triggers — “auto-edit multicam”, “做个剪辑”, “切几个机位”, “把这几个视频合成一个”, “combine these angles”, “PiP overlay”.
Use when the user has 2+ video or audio recordings of the same event captured on different devices (cameras, phones, separate audio recorders) and wants them aligned to a single common timeline. Outputs only a lightweight `.sync.json` sidecar per input; original files are never re-encoded. Triggers: "多机位同步", "对齐这几个机位", "match camera timelines", "sync these angles", "audio drift between cameras", "separate audio recorder", "Riverside / Zoom recording that needs to line up"
Derive height, tangent-space normal and packed MER (metalness, emissive, roughness) maps from an albedo/base-color texture for Blockbench PBR materials, using PyPBR (Python) or vgpu + Dawn WebGPU (Node.js), with optional Depth Anything V2 height estimation and MER inferred from the albedo plus the normal's curvature. Use after a color texture exists (painted, imported, or generated with blockbench-gpt-image-textures) when a material needs normal, heightmap or MER channels, a Bedrock RTX texture_set, or when asked to make a normal map, bump map, height map, roughness, metalness or emissive map from a texture. Triggers on albedo to normal, normal map from texture, height map from albedo, MER map, roughness from texture, metalness map, emissive map, Depth Anything, PyPBR, vgpu, Sobel normal, DirectX vs OpenGL normal, texture_set normal, heightmap or metalness_emissive_roughness layer.
Actually generate reference images for characters in any story (novel adaptations, your own original stories, or standalone character designs — no novel text needed): describe the character in a paragraph, break it into layered fields, confirm after filling them in, first generate one front-view full-body anchor image, and make all other views (headshot, 90° profile, back view, details, 45° headshot) reference only that anchor, generating images in tiers as needed. Every image carries an identifier and can be regenerated individually; after regeneration, images that are now outdated are marked automatically. Supports Qwen Image (ComfyUI), codex built-in image generation, OpenAI Images API (GPT Image 2) and custom commands, and the main image and derived images can use different models. Produces asset.json + a set of PNGs + a report you can open by double-clicking. Zero dependencies. If you only want prompts and not images, use novel-characters. Use when asked to 出角色图、做角色参考图、给我的角色出图、定妆照、真出三视图、character reference images.
CSS animation adapter patterns for FrameVideo. Use when authoring CSS keyframes, animation-delay based timing, animation-fill-mode, animation-play-state, or CSS-only motion that FrameVideo must seek deterministically during preview and rendering.
Asset preprocessing for FrameVideo compositions — text-to-speech narration, Chanjing background music and sound effect downloads, audio/video transcription (Whisper), and background removal for transparent overlays (u2net). Use when generating voiceover from text, downloading BGM or SFX, transcribing speech for captions, removing the background from a video or image to use as a transparent overlay, choosing a TTS voice or whisper model, or chaining these (TTS → transcribe → captions).
Generate professional voiceover narration for a video with audio-video sync using Azure TTS by default, or Gemini 3.1 Flash TTS when configured. Use this skill whenever the user wants to add narration, voiceover, commentary, or voice dubbing to any video file — even if they just say "add audio to this video" or "make a narrated version." Also trigger when the user has a screen recording, demo, tutorial, or presentation video that needs a voice track. Trigger on Chinese requests like "视频配音", "给视频加旁白", "录屏解说", "视频加语音", "视频添加声音", "生成视频旁白", "自动配音", "视频解说词".
Generate production subtitle artifacts from the final narration audio or final merged video using Volcengine Doubao ASR word timestamps. Use for local IndexTTS2 videos, Xiaohei page videos, talking-head delivery subtitles, SRT/VTT export, or whenever subtitle timing must match the final audio. Do not use script-length interpolation for final deliverables.
Share a short video that explains what you did. Use it for demos, walkthroughs, PR recaps, async handoffs, visual bug reports, design reviews, or any completed work that is easier to review in video form. Do not use it for trivial answers, active back-and-forth, unfinished work, or sensitive data.
Maintain the existing May Shorts teaching compositions using their legacy face-mode, scene-overlay, and karaoke-caption patterns. Use only when working on those examples or explicitly requesting this legacy scaffold. For new reels, Shorts, or short advertisements, use short-form-edit.
Transform a user-provided photo or image into a strange vintage Soviet or Eastern European children's book illustration with grotesque humorous cartoon energy, shaky ink, faded watercolor, dirty paper texture, awkward anatomy, nervous absurd expressions, sparse composition, and an absurd handwritten English rhyme. Use this skill when the user asks to turn a photo into an unsettling old children's book illustration, 1980s Eastern European illustration, weird Soviet cartoon book art, grotesque watercolor storybook art, or clumsy absurd illustrated caption style. Do NOT trigger for polished fantasy art, cute children's illustration, modern vector art, realistic portraits, anime, clean editorial illustration, or generic vintage filters.
Provider-independent still-image post-production skill for RAW/rendered intake, nondestructive development, exposure, white balance, tone, color, ICC proofing, masking, repair, compositing, truthful retouching by genre, generative disclosure, sharpening, noise, resampling, variants, metadata, export, QA, rights, and ethical finishing. Use when an agent must plan, perform, direct, or review still photography retouching and finishing for print, web, social, archive, publication, or client delivery; do not use for image-generation APIs, video grading, or campaign-specific production strategy.
Capture a website and create a FrameVideo 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.
Use when the user has a video + a target-language SRT and wants the video to actually speak that language — generates a time-aligned TTS voice dub. Routes by voice ID — Volcano (豆包) TTS for Chinese, edge-tts neural for any language. Defaults to one voice (single-speaker); opt-in multi-speaker via visual diarization. Outputs `*_<lang>_dub.mp4` with the dub audio in place of the original. Final mixing (audio bed + burn-in) is handed off to `/wjs-burning-subtitles`. Triggers — "配音", "中文配音", "Chinese dub", "voice over this", "dub the video", "TTS this SRT", "different voice for each speaker".
Create video compositions, animations, title cards, overlays, captions, voiceovers, audio-reactive visuals, and scene transitions in FrameVideo HTML. Use when asked to build any HTML-based video content, add captions or subtitles synced to audio, generate text-to-speech narration, create audio-reactive animation (beat sync, glow, pulse driven by music), add animated text highlighting (marker sweeps, hand-drawn circles, burst lines, scribble, sketchout), or add transitions between scenes (crossfades, wipes, reveals, shader transitions). Covers composition authoring, timing, media, and the full video production workflow. For dev-loop CLI commands (init, lint, inspect, preview, render) see the framevideo-cli skill; for asset preprocessing commands (tts, transcribe, remove-background) see the framevideo-media skill; for Chanjing digital humans, OAuth, and website-project synthesis see the chanjing-digital-human skill.
FrameVideo CLI dev loop — `npx framevideo` for scaffolding (init), validation (lint, inspect), preview, render, and environment troubleshooting (doctor, browser, info, upgrade). Use when running any of these commands or troubleshooting the FrameVideo build/render environment. For asset preprocessing commands (`tts`, `transcribe`, `remove-background`), invoke the `framevideo-media` skill instead.
Connect to Pascal and use its MCP tools to create, inspect, edit, validate, save, or hand off editable 3D building scenes. Use this skill whenever a user asks an agent to work in Pascal, make a room or building model, inspect a Pascal project, perform spatial edits, connect Pascal MCP, or return a verified Pascal editor link. It also governs safe local, existing-account, and explicitly authorized autonomous setup.
Production guidance for reusable layered 2D character rigs in browser-rendered and video media. Use for SVG hierarchies, pivots and constraints, FK/IK planning, pose and facial libraries, acting, cycles, deterministic handoff, crop variants, and rig QA; not for character concept continuity, 3D armatures, or motion-capture solving.
Anime.js adapter patterns for FrameVideo. Use when writing Anime.js animations or timelines inside FrameVideo compositions, registering animations on window.__fvAnime, making Anime.js seek-driven and deterministic, or translating Anime.js examples into render-safe FrameVideo HTML
Knowledge comic creator supporting multiple art styles and tones. Creates original educational comics with detailed panel layouts and sequential image generation. Use when user asks to create "知识漫画", "教育漫画", "biography comic", "tutorial comic", or "Logicomix-style comic".
AI image generation with OpenAI, Google, DashScope and Replicate APIs. Supports text-to-image, reference images, aspect ratios. Sequential by default; parallel generation available on request. Use when user asks to generate, create, or draw images.
Chanjing platform integration for FrameVideo: OAuth-aware digital-human video synthesis, public/custom digital-human and voice selection, Chanjing AI Creation image/video assets, platform BGM/SFX downloads, polling/downloading generated assets, and wiring Chanjing assets into FrameVideo compositions. Use when listing Chanjing people or voices, checking Chanjing auth in a platform workflow, submitting/polling digital-human videos, planning/submitting/downloading AI Creation image/video tasks, downloading Chanjing music or sound effects, or integrating Chanjing-generated assets into FrameVideo. For local Kokoro TTS/transcribe/remove-background use framevideo-media; for composition HTML conventions use framevideo
Author a new FrameVideo registry block (caption style, VFX block, transition, lower third) or component (text effect, overlay, snippet) and ship it as an upstream PR to the framevideo repo. Use ONLY when the user wants to CONTRIBUTE to the public catalog — for in-project caption/transition authoring use the `framevideo` skill, for installing existing registry items use the `framevideo-registry` skill
Complete color manipulation and green screen effects system. PROACTIVELY activate for: (1) Green screen/chromakey removal, (2) Color grading with LUTs, (3) Color correction (levels, curves, white balance), (4) Colorkey/color removal effects, (5) Hue/saturation manipulation, (6) Color space conversions (BT.709, BT.2020, HDR), (7) Color isolation effects, (8) Teal and orange look, (9) Vintage/film looks, (10) Color keying for transparency. Provides: chromakey and colorkey filters, LUT application (lut3d), curves and levels adjustment, color balance, selective color manipulation, color space handling, HDR tone mapping, professional color grading chains.
Draws fun, sharp isometric stickers (flat fill, thick black outline, extruded slab side) and animates them on a library of After Effects grade keyframe presets, replayed exactly the way lottie-web plays them, so the motion is fluid character animation and never a made up wobble. Any palette (six named sets with Riso by default, two brand colours, one colour, hue shift, custom slots) and four facing directions. Produces icon modules that play live on canvas in any web app (a React component ships with the kit) and export to looping alpha video (webm, HEVC mp4) with a poster. Use for "/fun-isometric", "animated sticker icons", "fun isometric icons", "Lottie style icons", "onboarding illustrations that move", or any request for playful animated icons, empty state art or feature stickers.
Lottie and dotLottie adapter patterns for FrameVideo. Use when embedding lottie-web JSON animations, .lottie files, @lottiefiles/dotlottie-web players, registering instances on window.__fvLottie, or making After Effects exports deterministic in FrameVideo
Expert-level media production, content management, streaming, broadcasting, and media technology systems. Use when the user mentions video, streaming, broadcast, CMS, or production, or when the task involves Media Production, Streaming and Broadcasting, Technologies, or Standards and Protocols.
Standalone multi-agent image generation skill for Hermes. Includes an internal design compiler, GPT-Image-2 generation via apimart.ai, case library reuse, interactive reference selection, batch workflows, and style-consistent series generation.
Best practices and techniques for writing effective AI video generation prompts. Covers: Veo, Seedance, Wan, Grok, Kling, Runway, Pika, Sora prompting strategies. Learn: shot types, camera movements, lighting, pacing, style keywords, negative prompts. Use for: improving video quality, getting consistent results, professional video prompts. Triggers: video prompt, how to prompt video, veo prompts, video generation tips, better ai video, video prompt engineering, video prompt guide, video prompt template, ai video tips, video prompt best practices, video prompt examples, cinematography prompts
Universal watermark removal with ML-based inpainting and automatic detection. Works on ANY watermark type (Google SynthID, Midjourney, DALL-E, stock photos, logos). Four methods: inpaint (ML, best quality), aggressive (fast), crop (fastest), paint (basic). Auto-detects watermark location in any corner. Use when: (1) Removing ANY type of watermark, (2) Google AI/Imagen/Gemini watermarks, (3) Stock photo watermarks, (4) Logo overlays, (5) Cleaning images for production, (6) Batch processing, or (7) User mentions 'watermark', 'remove watermark', 'clean image', 'SynthID'
Turn any topic into a narrated motion-graphics explainer video in Chinese or English, on a black canvas with a star-field or dot-field backdrop, TTS voiceover, subtitles and a chapter progress bar, every frame drawn in code with Remotion. Use when the user asks for an explainer, educational or science-communication video about a topic, or wants an article or document turned into a video. Includes compilable templates, a primitive library, voiceover/storyboard/render tools, style and motion specs, a multi-agent division-of-labor protocol and QC criteria, plus a complete sample video ("RAG and Knowledge Bases") as a quality benchmark.
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.