Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Use when one approved Layer asset must ship in several languages: dubbing a trailer or ad, voice-over in another language, burned-in subtitles or captions per market, or translating text that appears in an image, end card, store screenshot, or UI. Also when a dub drifts out of sync, translated copy no longer fits, or a font shows empty boxes for another script. Keywords: localisation, localization, dubbing, translation, subtitles, captions, voice-over, multilingual, markets, CJK, RTL
Use when one Layer video edit must be rendered many times with different media: an ad template filled per hook, per language, per placement or per character, a reusable intro or end-card template, or a batch of variants from one timeline. Also when a voice-over or clip should set the edit's length, or when a re-render must update an existing board timeline instead of adding a copy. Keywords: video template, slots, variants, batch render, auto length, reusable edit
Use when making store listing art for a game or app with Layer: an app icon and icon variants, a screenshot set with captions, a feature graphic, Steam capsules and header art, or an app preview video. Also when one key art must become every size a store asks for, or a logo or icon stops reading at thumbnail size. Keywords: app icon, store listing, screenshots, App Store, Google Play, Steam capsule, feature graphic, ASO, preview video
Use when a character or presenter on Layer has to speak on camera: a talking-head presenter, a UGC-style ad, a character delivering dialogue in a cutscene or trailer, or lip-syncing existing footage to a new line. Also when the mouth drifts out of sync, a line overruns the clip, or a voice changes between shots. Keywords: lip sync, lipsync, talking head, presenter, UGC, spokesperson, dialogue, voice over, avatar, talking character
Use when directing a single generated shot on Layer: choosing the camera move, lens, framing, lighting and pacing, applying a camera preset such as a dolly zoom or orbit, or timing a performance to the clip length. Also when a shot drifts, moves the wrong way, or feels flat or rushed. Keywords: cinematography, camera move, dolly, orbit, crane, tilt, FPV, handheld, lens, framing, lighting, shot language, technique preset
Use when giving a Layer video cut one consistent finish: a colour wash in brand colours, a retro CRT or VHS look, film grain and light leaks, a glitch hit on a cut, keying out a green or blue screen, lifting a cut-out logo off the frame, or filling the bars behind a clip. Also when shots in one edit do not match. Keywords: look, grade, filter, tint, duotone, film grain, vignette, scanlines, glitch, chroma key, green screen, drop shadow
Edit images with Google Nano Banana 2 (image-to-image edit endpoint) on RunComfy. Documents Nano Banana Edit's strengths (preserve subject identity, swap background, localize edits with spatial language, multi-image batch edits up to 20 inputs), the schema, and when to route to GPT Image 2 edit / Flux Kontext / Nano Banana 2 t2i instead. Calls `runcomfy run google/nano-banana-2/edit` through the local RunComfy CLI. Triggers on "nano banana edit", "edit with nano banana", "image edit nano banana", or any explicit ask to edit with this model
Create infographic prompts and render workflows with Nano Banana 2.1, with adaptable low-noise presets for posts, decks, reports, and explainers. Trigger on infographic, Nano Banana, Gemini image, executive visual, blog diagram, or presentation visual. Do NOT use for logos, memes, raw dashboards, or non-infographic image edits.
Designing scientific schematics, diagrams, and graphical abstracts. Covers tool selection (BioRender, Inkscape, Affinity, PowerPoint), design principles for pathway diagrams, mechanism schematics, experimental workflows, and journal graphical abstracts. Includes composition, icon sourcing, color for biological entities, and accessibility. Use when creating illustrative (not data-driven) scientific figures.
Provider-independent workflow for producing terminal, IDE, documentation, browser, desktop, and application demonstration videos. Use when choosing authentic capture, browser automation, synthetic UI, or hybrid treatment; scripting actions and narration; resetting demo state; protecting secrets and personal data; directing cursor/callouts; creating crop variants; and reviewing workflow truth, readability, accessibility, and provenance.
Turns a plain-text script into an AI-narrated voiceover, choosing delivery through voice selection and script craft rather than assuming a literal emotion or speed dial exists. Works standalone (no video yet, narration lands on a minimal carrier project) or attached to an existing project's clips. Use when the user says "generate a voiceover from this script", "I need narration with an energetic tone", "turn this script into AI narration", "make this voiceover sound warmer and slower", or "read this script out loud for my video".
Plan, generate, QA, and package continuous raster animation frame sets, GIF previews, and transparent spritesheet atlases. Use when the user asks for a GIF animation, animation frames, sprite rows, web/game animation assets, or reusable visual animation materials.
Author production procedural materials in Three.js. Use for simulated paper burning with coupled sheet combustion and pyro fluids, luminescent jellyfish with simulated appendages and layered tissue optics, luminescent butterflies with articulated scale glints and finite wing emitters, hybrid texture-backed PBR soil and moss with procedural displacement and masks, upward-facing model moss accumulation, atlas filtering, specular AA, planet-space fields, terrain wetness, deformable granular sand with mass-coupled grain shading, simulated tearable denim with implicit yarn mechanics, wear-driven failure and frayed fibres, woven cloth with GPU XPBD mechanics and yarn microstructure, simulated fur with interactive grooming and strand shading, lava and emissive surfaces, reflective wave-optical diffraction gratings, air-film-air soap bubbles with Airy interference, raytraced diamond and gem refraction with internal reflection and dispersion, image-space glass transmission with spectral dispersion and volume absorption, deforming softbody jelly materials with XPBD mechanics and refractive caustics, per-instance dissolve, authored PBR identities, derivative normals, and custom direct-light shadow modulation
Implement text-to-speech (TTS) capabilities using the z-ai-web-dev-sdk. Use this skill when the user needs to convert text into natural-sounding speech, create audio content, build voice-enabled applications, or generate spoken audio files. Supports multiple voices, adjustable speed, and various audio formats.
Select, integrate, and operate multi-model video-generation gateways — hosted inference aggregators such as fal.ai, Replicate, and WaveSpeed that expose many third-party video (and video-adjacent audio) models behind one account, one API surface, and one bill. Use when deciding whether to route video generation through a gateway versus a direct model-provider API; when comparing gateways on catalog, async/queue job design, webhooks, pricing, schema discovery, and version pinning; and when operating gateway workloads in production — spend controls, retries, cross-gateway failover, content-safety differences, data retention, and commercial-rights passthrough of the underlying model. Not for direct model-provider APIs, local/self-hosted inference, model training/fine-tuning, or image-only gateway use.
Make a 30-45 second enrollment-driving trailer for an academy or LMS course - who it's for, three fast "you'll learn" highlights, the outcome promise, and an enroll CTA, with energetic branded motion scenes interleaved with brief product peeks. Use when the user says "make a course trailer", "promo video for our academy course", "enrollment video", "teaser for this course", or "advertise this course to our customers".
Generate, inpaint, and outpaint music with ACE Step on RunComfy via the `runcomfy` CLI. ACE Step is StepFun-AI's open-weights music foundation model — tag-driven composition (genre, mood, instruments), multilingual lyrics with section markers, 5 s to 4 min stereo output, $0.0002–0.0003 per second (≈ 27× cheaper than ElevenLabs Music). Four endpoints: ACE Step text-to-audio (the default), ACE Step 1.5 text-to-audio (50+ language lyrics, refined structured-lyric handling), ACE Step audio-inpaint (regenerate a time range inside an existing track), ACE Step audio-outpaint (extend an existing track before or after). Triggers on "ace step", "ace-step", "acestep", "ACE music", "open music model", "cheap AI music", "inpaint audio", "audio inpaint", "extend music", "audio outpaint", "lengthen track", "music with tags", or any explicit ask to generate or edit music with ACE Step.
Append one branded end card to an existing video with a single clear next action - a link, trial, docs page, or contact - and a clean close. Use when the user says "add a CTA at the end", "add an end card", "close the video with a call to action", "add an outro with our link", or "end this with 'start your trial'".
Add a persistent, unobtrusive corner logo watermark to an existing video - one corner, low opacity, present for the full runtime without covering any action. Use when the user says "add our logo to this video", "watermark this video", "put our logo in the corner", or "brand this video with our logo".
Generate AI images with FLUX, Gemini, Grok, Seedream, Reve and 50+ models via inference.sh CLI. Models: FLUX Dev LoRA, FLUX.2 Klein LoRA, Gemini 3 Pro Image, Grok Imagine, Seedream 4.5, Reve, ImagineArt. Capabilities: text-to-image, image-to-image, inpainting, LoRA, image editing, upscaling, text rendering. Use for: AI art, product mockups, concept art, social media graphics, marketing visuals, illustrations. Triggers: flux, image generation, ai image, text to image, stable diffusion, generate image, ai art, midjourney alternative, dall-e alternative, text2img, t2i, image generator, ai picture, create image with ai, generative ai, ai illustration, grok image, gemini image
Annotate every screenshot in a help article - zoom, crop, blur, and arrows - so each image points at exactly the control its step describes. Use when the user says "annotate the screenshots", "add arrows to the images in this article", "zoom the screenshots in on the right buttons", "blur the sensitive bits in the doc images", or "make the article's screenshots clearer".
Convert MP3/WAV/FLAC audio files to MIDI (.mid) and MusicXML (.musicxml) with full music analysis. Two engines: Basic Pitch (general-purpose polyphonic) and Piano Model (high-accuracy piano, 96.7% F1). Optional Demucs stem separation. Use --engine piano for piano/keyboard music for best results. Trigger when the user wants to: transcribe audio to MIDI, convert music to sheet music/notes, extract notes from audio, get MusicXML from a recording, analyze tempo/key/chords of a song, separate stems (vocals/drums/bass) and transcribe each, or any audio-to-notation task. Also triggers for requests like "convert this MP3 to MIDI", "get the notes from this song", "what key is this song in", "transcribe this audio", "separate and transcribe stems".
Time of day and atmosphere in Blender 5.2 - measured sky + sun for dawn, morning, noon, golden hour, sunset, blue hour, overcast, night/moonlight with stars and city glow, exposure balance, height fog from visibility, god rays, Cycles vs EEVEE volumes, compositor grade, stylized moods (cyberpunk neon, horror, underwater, desert). Use for any day/night, sunset, foggy or mood lighting scene.
Adapts one base product demo into several vertical-specific variants (healthcare, finance, retail, or whatever segments the deal calls for) by analyzing the base demo's structure, working out how each vertical's real pain points map onto existing moments, then rewriting narration and on-screen text so the why-this-matters framing genuinely shifts per audience, not just the terminology. Each vertical is built on its own duplicated copy so the base demo stays intact. Use when the user says "make versions of this demo for healthcare, finance, and retail", "adapt this demo for different industries", "create vertical-specific variants of our demo", or "reframe this recording for three industries".
Condenses one full product demo into a single, continuous, social-ready teaser hitting a specific target runtime (default 60 seconds). This is compression, not selection: the whole demo's narrative arc (setup, each capability, the payoff) gets tightened into one coherent story, cutting dead air and repetition aggressively while preserving flow from beat to beat, rather than lifting out several separate highlight moments. Use when the user says "cut this demo down to 60 seconds", "make a short teaser from this full demo", "condense this into a social-ready clip", "give me a 30-second version of this demo", or "shrink this walkthrough into one tight clip for social".
Produce, direct, integrate, and quality-control ElevenLabs text-to-speech for narration, character performance, multilingual media, long-form voiceover, and streamed or batch AI-content audio. Use when selecting ElevenLabs models or voices; preparing spoken text; controlling pronunciation, pacing, emotion, and code-switching; using TTS APIs; cloning or designing a voice with consent; repairing artifacts; mastering deliverables; or evaluating generated speech. Excludes music, sound-effect generation, conversational-agent design, and general speech recognition except transcription used to verify TTS.
Edit existing videos using AI — remix style, edit content, upscale resolution, remove background, or add audio/sound effects. Use when the user requests "Edit video", "Remix video", "Upscale video", "Remove video background", "Add sound to video", "Video to audio".
Turn release notes or a changelog entry into a tiered feature announcement video - a 60-90s big-launch cut with an animated hook, demo, and CTA, or a 20-30s quick-hit for minor releases - built around real demo footage of the feature. Use when the user says "make a feature announcement video", "we're shipping X, make a release video", "turn these release notes into a video", "changelog video", or "launch video for this new feature".
Generate images with Flux 2 Klein (Black Forest Labs' distilled fast variant of Flux 2) on RunComfy — bundled with the model's documented prompting patterns so the skill gets sharper output than naive prompting against the same model. Documents Flux 2 Klein's strengths (sub-second latency, multi-reference brand styling, declarative subject-first prompts), the step-count strategy (4–8 for fast iteration, ~25 for polish), the 9B vs 4B variant trade-off, and when to route to Flux 2 Pro / Seedream 5 / GPT Image 2 instead. Calls `runcomfy run blackforestlabs/flux-2-klein/9b/text-to-image` (or `/4b/`) through the local RunComfy CLI. Triggers on "flux 2 klein", "flux-2-klein", "flux klein", "BFL flux 2", or any explicit ask to generate with this model
Design, structure, and animate code-first video compositions using the framefields WebGPU SDK. Use when creating scenes, layout trees, kinetic typography, camera trajectories, vector shapes, or audio-synchronized motion graphics in framefields.
Invoke for guided game design document authoring -- section-by-section GDD creation with validation, pillar alignment, and scope tier marking. Triggers on: "write GDD", "create GDD", "game design document", "document my game", "GDD authoring", "fill in GDD". Do NOT invoke for reviewing an existing GDD (use game-design-review) or designing specific game systems (use game-designer). Part of the AlterLab GameForge collection.
Plan, generate, edit, and quality-check still images with Google's Gemini native image models (Nano Banana), including Gemini 3.1 Flash Image, Gemini 3.1 Flash-Lite Image, Gemini 3 Pro Image, and legacy Gemini 2.5 Flash Image. Use for model and API-route selection, conversational editing, multi-reference composition, thinking and Google Search grounding, resolution/aspect-ratio control, exact-text and brand workflows, SynthID/provenance, safety, privacy, rights, and migration from legacy Gemini or Imagen image endpoints. Do not use for Veo/video generation or Gemini Live/audio work.
Make every click in a screen recording legible: add a brief visual emphasis on the cursor at each click and selection so viewers never lose the pointer. Use when the user says "highlight my clicks", "make the cursor visible", "viewers can't see where I'm clicking", "add click effects", or "emphasize the mouse actions".
Automatically detects grid split lines and splits a single multi-panel image into multiple separate images. Triggers when the user says four-grid split, nine-grid split, cut into four, 2x2/3x3 cut, separate grid image, split a collage, or separate grid panels.
Brightens shadows and recovers detail in existing images that are dark, night-time or underexposed, while keeping highlights from clipping. Trigger only when the user explicitly states it is low light, night-time or underexposure repair and has provided an image; a general "brighten" or ordinary colour-grading request does not trigger it on its own.
Image geometric transforms, smart composition and size changes (without altering image content), covering rotation/flipping, magnification by a factor, lossless downscaling, resizing, smart composition, and canvas extension/outpainting. Triggers when the user mentions rotate, flip, mirror, enlarge N times, shrink to xxx, outpaint, change to 16:9, portrait to landscape, outpaint, smart composition, recompose, resize.
A profile-driven coaching assistant for building an IP. On first use, create a private profile; afterwards, combine the profile, historical data and current evidence to re-diagnose the real problem, then work on persona and account positioning, content pillars, topic sourcing, evaluation and revision, short-video scripts and voiceover copy, on-camera performance intent, account launch and follower growth, series and account memory, content retrospectives, or content monetization. Also used for entry requests such as "新手入门", "怎么用" and "更新 ip-strategist". Natural language or the host-supported /ip-strategist both route through the same entry point.
Use when making motion graphics with Layer: a logo reveal or sting, an idle logo loop, animated titles, lower thirds, an end card or CTA card, kinetic type, or a branded intro or outro for a trailer or ad. Also when a reveal lands on a distorted logo, when generated lettering comes back misspelled, or when text must move without keyframes. Keywords: motion graphics, logo reveal, logo animation, sting, title card, lower third, end card, kinetic typography, intro, outro
Use when animating a hand-painted, vector-style or HD 2D game character with Layer: idle, walk, run, attack, hit or death cycles, 4-direction or 8-direction sprites for a top-down or isometric game, or a character sprite sheet from a single approved design. Also when frames drift off-model, feet slide, or directions do not match. Keywords: sprite animation, animation cycle, walk cycle, idle, attack, 8-direction, 4-direction, turnaround, sprite sheet, flipbook, top-down, isometric
Use when making game tiles with Layer that must snap together on a grid: isometric tiles, hex tiles, diamond or top-down map tiles, buildings and props with a footprint, the same building from four facings, or a tileset for Tiled, Godot or Unity. Also when generated tiles do not line up, stand at different heights, or light from opposite sides. Keywords: isometric, iso, hex, hexagon, tileset, tile map, dimetric, footprint, facing, Tiled, TileSet, strategy map, city builder
Use when exact words must appear on a Layer asset: a title, logo lockup, CTA, button label, price, or caption set in a brand font, or text that must be spelled exactly and stay editable. Also when generated text comes back misspelled, when a brand .ttf or .otf must be used, or when text already in an image must be changed. Keywords: typography, text overlay, brand font, lettering, title card, CTA, label, render text, compose, font file
Use when making user acquisition or performance marketing creative for a game with Layer: static ad concepts, hook variants, statics turned into motion, video ads for TikTok, Meta, YouTube or ad networks, end cards and CTAs, or one master resized for every placement. Also when a creative test needs attributable variants. Keywords: UA, user acquisition, ad creative, performance ads, hook, CTA, end card, static to motion, variants, A/B test, placements, 9:16
Use when generating game visual effects with Layer: spell and skill effects, hit impacts, explosions, pickups and power-ups, loot chest and reward openings, auras, portals, or looping ambient effects such as fire, smoke and magic swirls. Also when an effect comes back with a scene around it, cannot be blended in the engine, or plays with no wind-up. Keywords: VFX, FX, spell effect, impact, explosion, particle, flipbook, aura, loot chest, additive, sprite sheet
Use when a Layer video has more than one shot and they must cut together: a trailer, ad, cutscene or short story where the same character, set and motion carry across cuts. Also when a character changes face or costume between shots, when shot two does not start where shot one ended, or when drafting a whole sequence before paying for finals. Keywords: multi-shot, continuity, sequence, shot list, character reference, elements, next shot, Seedance, cutscene
Use when a generated or supplied clip on Layer is nearly right and should be repaired, not rerolled: extending it, inserting into it, swapping an object or a character, transferring motion, relighting, smoothing motion or making slow motion, upscaling, denoising, deblurring, or converting to HDR. Keywords: fix video, extend clip, swap character, motion transfer, relight, interpolation, slow motion, frame rate, upscale video, denoise, deblur, HDR
Memory-driven AI visual assistant. Supports 7 core capabilities (image generation, editing, face swap, virtual try-on, beauty enhance, image-to-video, motion transfer) and 17 scenario workflows. Triggered only when the user explicitly asks for a supported personalized visual workflow such as avatar series, background swap, try-on, style remix, daily card, or bring-image-to-life. It reads Meitu credentials, may read local profile/memory files, invokes the local `meitu` CLI, may send selected local context to Meitu OpenAPI, and writes outputs plus optional memory/profile updates locally.
Produce 3D assets with Meshy (meshy.ai) through its REST API and web platform — text-to-3D, image-to-3D, multi-image-to-3D, AI texturing/PBR, remesh/topology control, UV, and auto-rigging/animation. Use this skill when an agent must generate a mesh from a prompt or reference image, control polygon count and topology, apply or restyle textures, rig a humanoid for animation, export to GLB/FBX/OBJ/USDZ/STL for Unity/Unreal/Blender, drive Meshy's asynchronous job API (auth, polling, streaming, webhooks, credits, rate limits), review generated meshes for game/film/print readiness, or decide whether Meshy is the right tool at all. Not for editing an existing hand-modeled asset's geometry, hard-surface CAD with exact dimensions, or guaranteed-watertight manufacturing parts.
Programmatically generate Minecraft builds as vanilla structure NBT files and open Lodestone-powered previews. Use when an agent needs to create, export, inspect, or iterate on Minecraft structures, schematics, block palettes, or build recipes using JavaScript and .nbt output.
Generate, edit, composite, stream, and production-review still images with OpenAI GPT Image models. Use when an agent must choose between GPT Image 2 and legacy GPT Image/DALL-E integrations, select the Image API or Responses API, build prompts and reference-image workflows, use masks or multiple inputs, preserve identity or brand details, render in-image text, handle output files and partial images, estimate cost and rate limits, migrate deprecated image code, or apply OpenAI image safety, consent, privacy, provenance, and rights requirements. Do not use for OpenAI video or audio generation.
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.