Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Use Resemble AI Chatterbox for text-to-speech and voice-cloning workflows, including local open-weight Chatterbox, Chatterbox Multilingual, Chatterbox Turbo, and Resemble-hosted Chatterbox API routes. Apply when an agent must choose a Chatterbox variant, prepare reference-voice inputs, control emotion/paralinguistic delivery, plan multilingual TTS, handle PerTh watermarking, or make production decisions about consent, privacy, licensing, latency, and audio QA.
Master hook creation and audience retention for viral video success. PROACTIVELY activate for: (1) Video hook optimization, (2) Opening second strategies, (3) Audience retention improvement, (4) Attention capture techniques, (5) Psychological triggers, (6) Pattern interrupts, (7) Watch time optimization, (8) Drop-off prevention, (9) Engagement psychology. Provides: Hook formulas, psychological trigger frameworks, retention benchmarks, pattern interrupt techniques, and actionable retention improvement strategies.
Generate a complete video plan with optimized title, thumbnail, and hook concepts based on research. Orchestrates specialized skills (youtube-title, youtube-thumbnail, youtube-video-hook) to create production-ready video plans. Use after research is complete or when the user wants to plan a new video.
Refined AI short-drama IP creation skill (v0.6.0). Three-phase architecture: Phase 1 creation (script + ref images, iterated repeatedly) → Phase 1.5 storyboard images (1-4 candidate stills per grid = the video's first frame, the industrial-grade core layer) → Phase 2 production (unlocked per episode, automatically choosing among 4 models). Key v0.6.0 upgrades: ① iron rules for craft bias in the ref library (modern skyscrapers easily turn anthropomorphic, ancient architecture always yields chibi human bodies, so pitfalls must be avoided during the early IP design stage) ② Jimeng 5.0 failure modes + sensitive-word list (distinguishing 3 types of failures / hard 1500-character prompt limit / villain-word replacement table / ambiguous-word list) ③ 8-segment storyboard prompt template (CHARACTER/BACKGROUND/ACTION/SCENE/CAMERA/LIGHT/TEXT/STYLE) + the NOT humans clause must always be added. Carried over from v0.3.0: Phase 1.5 storyboard layer, 4-model video, ref 5-8 optimal. v0.2.0: bash → Python subprocess, 36 grid × 4-10s variations, the 7 tricks for guaranteed hits on Hongguo, an industrial-grade ref library of 80-150 images, 300-500 characters per prompt. Always trigger when the user mentions short dramas, micro short dramas, vertical-screen dramas, AI short dramas, AI comic dramas, script writing, storyboards, production with Jimeng/Seedance, IP adaptation for Hongguo/Fanqie/Douyin, wish-fulfillment dramas, rebirth, time travel, live-in son-in-law, wife-chasing, miracle doctors and diviners, AI comic-drama spectacle, or requests like "help me make a short drama".
Audits AI-generated or AI-enhanced listing images against the 2026 Amazon disclosure and representation policy before submission. A 5-gate flow catches the issues that trigger suppression or a misrepresentation flag. Use when a user asks about AI image policy, AI photo rules, image suppression, or 2026 image compliance. Trigger phrases: "AI image Amazon policy", "AI generated photos listing", "image suppression AI", "Amazon AI image rules 2026". Works with zero tools.
Automate Blender 5.2 at scale - headless CLI flags, batch convert/inventory/run-any-skill-script over folders of .blend/.fbx/.glb/.obj/.usd, sharding, bpy pip module, library linking, path management, render-farm and CI patterns. Use for batch jobs, format conversion, library maintenance and pipeline tooling.
Provider-independent production workflow for agents assembling, auditing, executing, and handing off ComfyUI node-graph workflows for image, video, upscale, inpaint, conditioning, and batch media generation; use when a task involves ComfyUI workflow JSON/API graphs, model and custom-node inventories, reproducibility, provenance, safety/rights review, local or cloud execution hygiene, or production QA.
Provides comprehensive guidance for DALL-E AI image generation including prompt creation, image generation, variations, and editing. Use when the user asks about DALL-E, needs to generate AI images, create image variations, or work with DALL-E API.
ElevenLabs AI music generation - create original music from text prompts via inference.sh CLI. Capabilities: text-to-music, custom duration up to 10 minutes, genre/mood/instrument control, royalty-free commercial use. Use for: background music, soundtracks, jingles, podcasts, video scores, game audio. Triggers: elevenlabs music, eleven labs music, ai music, generate music, music generation, compose music, ai composer, create song, soundtrack, background music, jingle, elevenlabs compose, music ai
ElevenLabs voice isolator - remove background noise and isolate vocals from audio via inference.sh CLI. Capabilities: noise removal, voice extraction, audio cleanup, background removal. Use for: podcast cleanup, interview audio, music vocals, noisy recordings, audio restoration. Triggers: voice isolator, noise removal, background removal, isolate voice, clean audio, remove background noise, audio cleanup, voice extraction, elevenlabs isolator, eleven labs noise, vocal isolation, denoise, audio restoration, voice separation
Generate images with FLUX models (Black Forest Labs) via inference.sh CLI. Models: FLUX Dev LoRA, FLUX.2 Klein LoRA with custom style adaptation. Capabilities: text-to-image, image-to-image, LoRA fine-tuning, custom styles. Triggers: flux, flux.2, flux dev, flux schnell, flux pro, black forest labs, flux image, flux ai, flux model, flux lora
Generate high-fidelity portraits from a reference image (people or pets), covering ID photos, professional headshots, artistic portraits, pet portraits, group portraits, hairstyle adjustments, and two-person composite shots. Trigger only when the user explicitly requests portrait generation, portrait stylization, or two-person composite photos.
Personal IP diagram creator. Use your IP character to generate bespoke content diagrams. It turns photos, profile screenshots, bios, account materials or character profiles that the user has authorized into a consistent minimalist hand-drawn diagram character, and then generates long-form article illustrations, knowledge cards for image-and-text accounts, phone posters, method breakdown diagrams, storyboard story diagrams, IP talk PPTs, livestream sharing decks and fully illustrated hand-drawn PPT pages. It emphasizes a knowledge-diagram system of IP protagonist + execution agent: the user's IP does the presenting, breaking down, annotating and dispatching, while the execution agent moves cards, gives feedback, tries and errors, flags risks and hands over results. Use this Skill when the user wants a personal IP character, wants to make knowledge diagrams with their own character, knowledge cards for image-and-text accounts, article illustrations, course diagrams, content posters, method breakdown diagrams, IP talk PPTs, livestream sharing decks, or uploads photos / profile screenshots / bios for content visuals.
Source authentic, high-res PUBLIC-DOMAIN artwork from museum open-access APIs (Met, Cleveland, SMK, Rijksmuseum, NGA, Art Institute of Chicago, Getty, Smithsonian) instead of AI-generated or generic-stock imagery. The default move whenever a visual needs an aesthetic, credible image (blog heroes, decks, social cards, essay/spec figures). Verified keyless recipes + licensing rules inside.
Use when writing or auditing talking-head / narrative / "yap" content — reel scripts, video scripts, story-driven posts, any piece where someone talks to camera or tells a story. The hook stops the scroll; storytelling holds the watch after it. Not for pure tactical demos or bare listicles with no through-line.
Use when the user needs an end-to-end article illustration workflow in this repository that preserves Type x Style planning, loads article-illustration preferences, recommends Alibaba Cloud image backends, and produces a Markdown article with inserted local image references.
Facial rigs and shape keys in Blender 5.2 — ARKit 52 blendshapes, visemes for lip-sync, sculpting expressions, splitting symmetric shapes into Left/Right, mirroring, corrective and combination shapes, drivers and control rigs, bone+blendshape hybrids, export to engines (Unity, Unreal, Roblox dynamic heads, VRM/web). Use for faces, expressions, lip-sync, or any shape-key work.
Sculpting in Blender 5.2 — base-to-detail workflow, voxel remesh, dyntopo, multires, 5.x brush types, masks/face sets, symmetry, plus headless-safe procedural ops for agents (remesh, symmetrize, multires, inflate, noise detail, smooth) and the sculpt-to-game handoff. Use for organic models, characters, creatures, rocks and bake sources.
Guide a user conversationally through generating a new ad video from a template video and a user-provided product image. The host agent performs video understanding, detailed analysis, and rewrite planning; BytePlus Seedance is used only for final video generation with local product images as generation references.
For Codex, Claude Code, WorkBuddy and other Agent Skills clients, turn a one-line idea, story outline, novel, copy, script, storyboard or existing material into understandable, connectable, publishable AI short dramas and AI short films. Suitable for landscape or portrait content on Douyin, Hongguo Short Drama, WeChat Channels, Kuaishou, Bilibili, YouTube and more; supports live action, standalone short films, anthology series, serial dramas, pets, anthropomorphic fantasy, animation/anime-style dramas, UGC and branded storylines. Used for story diagnosis, screenwriting, storyboarding, character/recurring prop consistency, model routing for Wan 3, Seedance 2.5/2.0/Fast, MiniMax H3 and others, 15–30 second long takes, native dialogue, reference audio, dubbing, paid budgets, continuity editing, physical quality checks and cold-audience acceptance, especially to avoid fragmented hard-cutting, random jump cuts, static talking-head dialogue, identity/costume/prop drift, text distortion, wrong motion direction, clipping, flat observational shots, lip-sync/voice mismatch, music restarts and incomprehensible plots.
ElevenLabs speech-to-text with Scribe models and forced alignment via inference.sh CLI. Models: Scribe v1/v2 (98%+ accuracy, 90+ languages). Capabilities: transcription, speaker diarization, audio event tagging, word-level timestamps, forced alignment, subtitle generation. Use for: meeting transcription, subtitles, podcast transcripts, lip-sync timing, karaoke. Triggers: elevenlabs stt, elevenlabs transcription, scribe, elevenlabs speech to text, forced alignment, word alignment, subtitle timing, diarization, speaker identification, audio event detection, eleven labs transcribe
ElevenLabs voice changer - transform any voice to a different voice while preserving speech content and emotion via inference.sh CLI. Models: eleven_multilingual_sts_v2 (70+ languages), eleven_english_sts_v2. Capabilities: speech-to-speech, voice transformation, accent change, voice disguise. Use for: content creation, voice acting, privacy, dubbing, character voices. Triggers: voice changer, speech to speech, voice transformation, change voice, voice swap, voice conversion, voice disguise, eleven labs voice changer, elevenlabs sts, transform voice, ai voice changer, voice modifier
Provider-independent explainer video production for AI agents creating educational, product, concept, nonprofit/public-service, onboarding, training, animated, mixed-media, and short social explainers. Use when an agent must turn a topic, brief, research corpus, product, process, policy, dataset, or complex idea into a factual, accessible, narrated, captioned, storyboarded, generated-media-ready explainer video with claim review, localization, pacing, variants, and QA.
Precisely separate the foreground subject from the background in an image. Only five subject types are supported — people, pets, products, icons, and seals — and the output is a transparent-background PNG. Trigger when the user explicitly asks for cutout, background removal, transparent background, or extraction of the above subjects; other subject types and white-background isolation are not supported.
Generate standard ID photos based on the user's frontal portrait and explicit ID photo specifications, with automatic outfit change, background replacement, and compliance adjustments. Trigger only when the user explicitly requests an ID photo, 一寸/二寸, passport/visa/ID card/driver's license and other standardized ID photos, and has already provided a portrait photo; generic requests for business photos, professional portraits, or glamour shots do not trigger it on their own.
Lip-sync a face to a specific audio track on RunComfy via the `runcomfy` CLI. Routes across ByteDance OmniHuman (audio-driven full-body avatar from a portrait + audio), Sync Labs sync v2 / Pro (state-of-the-art mouth sync onto a video), Kling lipsync (audio-to-video and text-to-video with synced speech), and Creatify lipsync. The skill picks the right endpoint for the user's actual intent — portrait still + audio (avatar-style), source video + audio (mouth-swap on existing footage), or generate-and-sync from a script. Triggers on "lip sync", "lipsync", "make this video speak", "match audio to mouth", "dub video", "sync lips to voice", "Sync Labs", "voiceover sync", or any explicit ask to drive a face's mouth from an audio track.
Generate multi-style four-grid sticker packs from photos uploaded by the user (built-in Q version, 3D clay, pixel art, and Emoji styles, with support for custom styles), split into 4 separate stickers, and optionally convert them into animated GIFs. Trigger when the user mentions 表情包, 贴纸, sticker pack, sticker, emoji pack, 生成贴纸, 做表情包, Q版贴纸, 大头贴.
Operate WorldEdit safely for Minecraft 26.x and 1.21.x selections, region edits, masks, schematics, brushes, terrain shaping, and rollback. Use for command-driven world editing, not plugin development or general server administration.
Generate images with Google Nano Banana 2 (Gemini-family flash-tier text-to-image) on RunComfy — bundled with the model's documented prompting patterns so the skill gets sharper output than naive prompting against the same model. Documents Nano Banana 2's strengths (rapid iteration, in-image typography rendering, predictable framing, optional web-grounded context), the resolution-tier pricing, the safety-tolerance dial, and when to route to Nano Banana Pro / GPT Image 2 / Flux 2 / Seedream instead. Calls `runcomfy run google/nano-banana-2/text-to-image` through the local RunComfy CLI. Triggers on "nano banana", "nano-banana-2", "nano banana 2", "google image gen", "gemini image", or any explicit ask to generate with this model
Generate images with Alibaba Qwen-Image-2.0-Pro via inference.sh CLI. Professional text rendering, fine-grained realism, enhanced semantic adherence. Ideal for posters, banners, and text-heavy designs. Triggers: qwen image pro, qwen-image-pro, qwen 2 pro, alibaba image pro, dashscope pro, professional text rendering
Generate and edit images with Alibaba Qwen-Image-2.0 models via inference.sh CLI. Models: Qwen-Image-2.0 (fast), Qwen-Image-2.0-Pro (professional text rendering). Capabilities: text-to-image, multi-image editing, complex text rendering. Triggers: qwen image, qwen-image, alibaba image, dashscope image, qwen image 2, qwen image pro
Analyze, recreate, align, and verify reference-video motion. Use when the user asks to recreate, copy, match, align, pixel-align, or compare a video; when a HyperFrames or Remotion remake is judged not aligned; when fixed reference playback and reusable motion components must be separated; or when delivery needs full-frame runtime, render, decode, boundary, timing, and archive evidence instead of sampled screenshots alone.
Generate SaaS product launch and software demo video prompts for Seedance 2.0 on Higgsfield. Use whenever the user wants a product launch video, app demo, software walkthrough, feature showcase, startup promo, tech product reveal, or any SaaS/software marketing video. Triggers on SaaS, app, software, product launch, demo, feature, startup, tech, UI, dashboard, landing page video.
Procedural noise functions in GLSL—Perlin, simplex, Worley/cellular, value noise, FBM (Fractal Brownian Motion), turbulence, and domain warping. Use when creating organic textures, terrain, clouds, water, fire, or any natural-looking procedural patterns.
Use meitu-cli to generate a brand-new image from scratch from a text description (no base image), covering general text-to-image, multi-image reference fusion and commercial-grade high-quality generation. Trigger when the user mentions generating an image, drawing an image, text-to-image, creating an image, concept art, promotional image, cover image, marketing image, product scene image, or movie-still style.
Extend or continue an existing video clip on RunComfy via the `runcomfy` CLI. Routes to Google Veo 3-1's `extend-video` and `fast/extend-video` endpoints — pick the source video plus a prompt describing what should happen next, and the model produces a clip that continues the original with consistent motion, lighting, and subject identity. Use when the user has a short Veo clip and wants it longer, or wants a chained narrative built shot-by-shot from a single seed clip. Triggers on "extend video", "continue video", "longer video", "video extend", "make this clip longer", "Veo extend", "chain video shots", "video continuation", or any explicit ask to take an existing video and add more frames after it.
Video outpainting on RunComfy via the `runcomfy` CLI — extend the spatial canvas of a video, change aspect ratio (9:16 vertical to 16:9 horizontal or vice versa), add environment beyond the original frame while preserving the central action. Routes prompt-shaped spatial extension through Wan 2-7 edit-video and points the agent at dedicated ComfyUI outpaint workflows when seam quality matters for hero delivery. Triggers on "video outpaint", "video outpainting", "extend video canvas", "expand video frame", "uncrop video", "aspect ratio change", "vertical to horizontal video", "16:9 from 9:16", "TikTok to YouTube", or any explicit ask to extend a video spatially beyond its original frame.
Generate instrumental music / BGM without vocals or lyrics, suitable for background music, atmosphere building and emotional rendering, 30–120 seconds long. Trigger only when the user explicitly asks to generate instrumental music/background music. During execution it reads Meitu credentials, calls the local `meitu` CLI, sends audio generation requests to the Meitu OpenAPI, and writes results to a local output directory.
Process and analyze file references uploaded by the user, extract key information, and provide application suggestions. Suitable for parsing @-symbol references and natural language references, and extracting information related to short drama planning
AI outfit swap that keeps the person's face and body shape and replaces only the clothing, supporting text descriptions or clothing reference images. Trigger only when the user explicitly asks to swap outfits, try on clothes, change clothes, or replace garments in an existing single-person image and has already provided the person's image.
Image sharpness enhancement / super-resolution upscaling, with the algorithm chosen automatically by content type (general / e-commerce product images / text documents). Triggered when the user mentions making something clearer, ultra-HD, HD restoration, quality improvement, super-resolution, upscaling to 4K/2K/1080P, old photo restoration, ultra-clear product images, ultra-clear documents, or making scanned files clearer.
Plan, prompt, generate, reference, edit, motion-control, and quality-review videos with Kuaishou's direct Kling AI video platform and API, especially Kling VIDEO 3.0, 3.0 Omni, 3.0 Turbo, native audio, multi-shot, elements, start/end frames, and image/video references. Use for first-party Kling video production or API integration, not Kolors image generation, avatars, standalone audio, consumer-only membership advice, or third-party Kling gateways.
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.