Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Produce presenter-led AI avatar videos with Synthesia Studio and Synthesia API. Use for Synthesia-specific avatar video planning, avatar/voice/language selection, scripted scenes, template personalization, API job lifecycle, localization/dubbing, brand/training/product workflows, consent/rights/safety checks, artifact custody, troubleshooting, and final QA.
Use for producing Tavus AI-human videos and real-time avatar conversations with Tavus Faces/Replicas, PALs/Personas, async Video Generation, CVI conversations, consent-safe likeness workflows, webhooks, backgrounds, localization, and avatar QA.
Generate and operate Tencent Hunyuan video through the managed TokenHub HY-Video-1.5 API or official local HunyuanVideo repositories. Use for text-to-video, image-to-video, hosted job lifecycle, pricing and region planning, local checkpoint selection, hardware and acceleration, prompt design, licensing, safety, provenance, and delivery QA.
Implement physically motivated sky and aerial-perspective systems in Three.js. Use for planetary atmospheres, ground-to-space transitions, Rayleigh/Mie scattering, precomputed LUTs, depth-based transmittance and inscattering, sun/moon discs, and atmosphere-aware lighting.
Create native TikTok video ads with Pexo — vertical 9:16, sound-on, hook-first. Give it your product and Pexo writes the script in TikTok's voice, sequences fast-paced shots, picks the models, and delivers a ready-to-post ad with trending-style music. Use for TikTok ads: "tiktok ad", "tiktok video ad", "ad for tiktok", "vertical video ad". NOT for landscape or long-form ads (use the video-ad skill).
Text-to-image generation and image editing via Together AI, including FLUX and Kontext models, LoRA-based styling, reference-image guidance, and local image downloads. Reach for it whenever the user wants to generate or edit images on Together AI rather than create videos or build text-only chat applications.
Build a sales training video straight from a topic, prompt, or outline, no recording needed. Sets an explicit learning objective the user confirms ("after this video, a rep should be able to X"), structures the script for retention (hook tied to quota and deals, core content mapped to real selling scenarios, an explicit recap close), and builds almost entirely from motion graphics. Use when the user says "make a training video on how to handle pricing objections", "build a sales training video from this outline", "train the team on our new positioning, no recording available", or asks to turn a topic into onboarding or enablement content for reps.
Complete YouTube long-form video optimization for maximum watch time, SEO, and viral potential. PROACTIVELY activate for: (1) YouTube video strategy, (2) Long-form content planning, (3) Watch time optimization, (4) YouTube SEO, (5) Thumbnail design strategy, (6) Audience retention techniques, (7) YouTube monetization, (8) Connected TV optimization, (9) YouTube algorithm understanding. Provides: Video length optimization, thumbnail best practices, SEO strategies, retention techniques, monetization guidelines, and analytics interpretation.
Plan, troubleshoot, and hand off provider-independent in-camera VFX and LED volume virtual-production work, with Unreal Engine/nDisplay/Live Link awareness. Use for ICVFX suitability, LED stage geometry, frustums, tracking, lens calibration, sync, render nodes, color, lighting/reflections, artifact mitigation, rehearsal, live-composite fallback, monitoring, records, QA, and handoff. Do not use for ordinary offline post compositing, generic storyboards, or full game production.
Production guidance for mainland-China Volcengine Doubao Speech text-to-speech. Use for TTS 1.0/2.0 selection, V3 bidirectional or unidirectional streaming, asynchronous long-text submit/query jobs, current voices and controls, timestamps and SSML caveats, expiring results, quota/error handling, authorized cloned voices, privacy, and audio QA. Do not use for international BytePlus endpoints.
Make YouTube Shorts that hold attention with Pexo — vertical 9:16, fast-paced, hook in the first 3 seconds. Describe your topic and Pexo writes the script, generates the shots, picks the models, and assembles a finished Short with music and captions. Use for YouTube Shorts: "youtube short", "make a youtube short", "shorts video", "short-form video". NOT for long-form or landscape video.
Create free Agnes video generation tasks and retrieve video results through Agnes-Video-V2.0, including text-to-video, image-to-video, multi-image video, keyframe animation, async polling, and video download. Use when the user asks for 免费视频, 免费的视频模型, free video API, Agnes video, Agnes-Video-V2.0, text-to-video, image-to-video, keyframes, or wants a free video model Agent skill.
Complete guide to FFmpeg filter_complex syntax, multi-input/output pipelines, and composition patterns. PROACTIVELY activate for: (1) filter_complex syntax questions, (2) Multi-input video composition (PiP, grids, overlays), (3) Video transitions (xfade, fades, wipes), (4) Audio mixing (amix, amerge), (5) Stream splitting and routing, (6) Named stream labels, (7) Complex filtergraphs, (8) GPU + CPU hybrid filter chains, (9) ABR encoding ladders, (10) Text and watermark overlays. Provides: Syntax reference, composition patterns, audio mixing examples, transition effects, GPU-accelerated filtergraphs, debugging tips.
Best practices for writing Remotion animations that stay intuitive for coding agents and remain editable in Remotion Studio Visual Mode — keeping values literal and inline where the Studio can find them, rather than hidden behind indirection. Use when writing or refactoring a Remotion composition, or when an animation is correct in code but not editable in the Studio. Triggers - Remotion Studio, Visual Mode, editable animation, interpolate, spring, composition props.
Make a complete video from a simple idea with Pexo. Describe what you want in plain words and Pexo handles everything — script, shots, model selection, music, subtitles, and final editing — and returns a publish-ready video. Use when the user just wants to make a video: "make a video", "make me a video about X", "create a video", "I need a video for ...". NOT for editing a video they already have.
Generate images and videos through the currently installed Media.io CLI. Use for text-to-image, image-to-image, text-to-video, image-to-video, reference-to-video and published workflows. Select the model from the bundled static catalog (`references/model-catalog.md`), which is generated from the production registry; only fall back to `mediaio model list` on the triggers listed in the discovery guardrail. Copy every `job_type` byte for byte — some contain a literal space, and display names often do not match the identifier. Effects may be used only when their parameters have been independently verified because the current CLI exposes `effect list` but not `effect get`. Always confirm the parameter schema with `model get` before submission. Submitting spends the user's credits, but the CLI stays quiet about the amount unless asked; surface the cost with `--show-credit`, and get an explicit approval first, only when the user is cost-sensitive or has raised credits, price or balance. When the balance runs short, or the user wants their task history or their uploaded files, hand over the product-page link printed by `mediaio link get` — never write or edit a media.io URL yourself. Use the human-readable discovery output, upload local files before generation, submit with `generate create`, wait with the separate `generate wait` command, and retrieve result files with `generate download` instead of reproducing signed result URLs. On hosts that sandbox local command networking, the first networked `mediaio` or `curl` Shell/Bash tool call must request host approval before process launch. Never probe network availability by running it in the default sandbox.
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 and edit images using Google Gemini image models. Use this skill when the user asks to create, generate, make, or edit images with AI. Supports text-to-image, image editing, style transfer, and multi-image composition. Trigger on requests like "create an image", "generate a picture", "make me a logo", "edit this photo", "add X to this image".
Cut one long video - a webinar, training session, or full walkthrough - into a series of short standalone videos, each with its own title and intro line. Use when the user says "split this into a series", "break this webinar into short videos", "make bite-sized videos out of this training", "turn this into separate videos per topic", or shares one long recording that should become several.
Build or transform lightweight realtime voice-chat apps with a talking character avatar from either a user-supplied photograph or a text description. Use when Codex is asked to create a talking avatar, photo avatar, character chatbot, roleplaying voice companion, mouth-sprite animation, or reusable OpenAI Realtime avatar app in Vite/Next.js, including BYOK API-key intake, identity-consistent image generation, audio-driven lip sync, testing, and deployment.
Use when implementing motion design, timeline animations, visual animation editors, animating Three.js/R3F scenes, creating keyframe animations, or using Theatre.js, @theatre/core, @theatre/studio, @theatre/r3f, theatric, or building animation tooling for the web.
Build, extend, or polish original 3D browser games and interactive prototypes with Three.js. Use when a user wants a playable WebGL game, a game-like Three.js experience, a 2D concept translated into 3D, or an existing Three.js project given a complete loop, controls, camera, collisions, UI states, audio, persistence, performance work, or production-build verification. Do not trigger for ordinary 3D product viewers, data visualizations, or decorative background scenes unless the request includes game mechanics or player interaction.
Build a deliberate final-image pipeline for advanced Three.js scenes. Use for depth, normal, albedo, and history ownership; GTAO or bent normals; bloom; eye adaptation; tone mapping; 3D LUT grading; effect-local render targets; and pass diagnostics.
Orchestrate Vozo CLI project workflows—single-tool create-to-download, multi-step pipelines, batch multi-language creation, and filtered bulk downloads. Use for translate dub, translate subtitles, visual translate, lipsync, folder/URL sources, and artifact downloads. Also use when the user asks what Vozo or this CLI is or what it can do (product introduction).
Build or transform ecommerce websites from loose company, product, or brand briefs into unconventional, brand-specific shopping experiences with original ImageGen product imagery. Use when a user asks for a crazy, wild, experimental, unconventional, anti-template, art-directed, or highly memorable online store; wants Codex to invent an ecommerce visual direction from business details; or wants a working storefront that combines creative concepting, generated product photography, responsive implementation, cart interactions, validation, and optional publishing.
Creative-writing session lead and entry point. Activate this to get a creative partner that brainstorms, drafts, critiques, revises, and maintains your story's knowledge base. Built for Claude.ai, where there are no subagents: it runs every mode of the work in one conversation. Load alongside the craft skills (writing-principles, prose-writing, scene-construction, prose-critique). Use at the start of any story session.
Felo SuperAgent API: AI conversation with real-time SSE streaming on a persistent LiveDoc canvas. Use when users want SuperAgent chat, continuous conversation, logo/branding design, or e-commerce product images. Do NOT use for tweet/X post writing — use felo-twitter-writer instead. Explicit commands: /felo-superAgent.
Complete FFmpeg + OpenCV + Python integration guide for video processing pipelines. PROACTIVELY activate for: (1) FFmpeg to OpenCV frame handoff, (2) cv2.VideoCapture vs ffmpeg subprocess, (3) BGR/RGB color format conversion gotchas, (4) Frame dimension order img[y,x] vs img[x,y], (5) ffmpegcv GPU-accelerated video I/O, (6) VidGear multi-threaded streaming, (7) Decord batch video loading for ML, (8) PyAV frame-level processing, (9) Audio stream preservation with video filters, (10) Memory-efficient frame generators, (11) OpenCV + FFmpeg + Modal parallel processing, (12) Pipe frames between FFmpeg and OpenCV. Provides: Color format conversion patterns, coordinate system gotchas, library selection guide, memory management, subprocess pipe patterns, GPU-accelerated alternatives to cv2.VideoCapture. Ensures: Correct integration between FFmpeg and OpenCV without color/coordinate bugs. See also: ffmpeg-python-integration-reference for type-safe parameter mappings.
Invoke when the user asks about game mechanics, core loop, balance, progression, economy design, reward systems, onboarding, game feel, systems design, or GDD authoring. Triggers on: "mechanics", "core loop", "balance", "progression", "reward", "onboarding", "game feel", "systems design", "GDD". Do NOT invoke for creative vision (use game-creative-director) or economy monetization (use game-economy-designer). Part of the AlterLab GameForge collection.
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.
Generate videos using Google Veo models via the nano-banana CLI. Use this skill when the user asks to create, generate, animate, or produce videos with AI. Supports text-to-video, image-to-video animation, dialogue with lip-sync, and scene extensions. Trigger on requests like "create a video", "animate this image", "make a video clip", "generate footage", "produce a short film", "add motion to this".
Use when writing, planning, continuing, repairing, revising, or running marathon/"疯狂写作"/auto continuation mode for Chinese long-form fiction with recurring characters, multiple plotlines, persistent world rules, chapter-by-chapter continuity needs, or style-specific constraints.
Make a SaaS demo or explainer video with Pexo. Describe your software (or paste your app URL or a screenshot) and Pexo writes the script, generates the visuals, picks the models, adds narration and captions, and returns a finished, publish-ready video — for your landing page, onboarding, or sales. Use for SaaS and software videos: "saas video", "saas demo video", "software demo", "app explainer video". NOT for live screen recording — Pexo generates the visuals.
Build, repair, or continue a complete storyboard workflow from script to clean-board guidance. Use when the user wants story-first boarding grounded in FloobyNooby-style planning, clarity, staging, shot progression, motivated cuts, continuity, rough thumbnails, animatic review, structural revision, key panels, full rough board, and clean-pass rules rather than an analysis-heavy shot list.
Create professional SVG logos from concept briefs or descriptions. Use when generating SVG logo files, creating logo variations (horizontal, vertical, icon-only), or implementing logo designs. Triggers on "create SVG logo," "generate logo," "make a logo," "logo SVG," "design a logo," or when given a logo concept brief from logo-ideation.
Implement a production GTAO path in Three.js. Use for half-resolution horizon sampling, reversed-depth reconstruction, bent-normal encoding, full-resolution bilateral reconstruction, environment-light application, contact grounding, and halo diagnosis.
Turns a support agent's ticket-resolution recording (e.g. a screen-share where they walked a customer through fixing an issue) into a clean, reusable how-to video that any future customer with the same issue can watch. Strips customer names, account-specific screen data, ticket numbers, and small talk, then reframes the narration around the general problem and solution rather than one customer's specific case. Use when the user says "turn this ticket resolution into a how-to video", "make this support recording reusable for other customers", "generalize this screen-share into a customer-facing clip", or "turn this ticket recording into a help video".
Produces a batch of short, deflection-style how-to videos from a list of top recurring support ticket topics: one focused video per topic, each built from whatever source material actually exists for that topic (a real resolution recording, a written description, or steps supplied on request). Aimed at cutting repeat ticket volume by giving customers a self-serve fix before they file again. Use when the user says "make how-to videos for our top 10 recurring tickets", "batch-produce deflection videos from this list of common issues", "turn these frequent support topics into videos", or hands over a list of recurring ticket topics and asks for videos against each.
Use when editing images with Alibaba Cloud Model Studio Qwen Image Edit models (qwen-image-edit, qwen-image-edit-plus, qwen-image-edit-max, qwen-image-2.0 series and snapshots). Use when modifying existing images (inpaint, replace, style transfer, local edits), preserving subject consistency, or documenting image edit request/response mappings.
Production guidance for international BytePlus Seed Speech text-to-speech. Use for selecting TTS 1.0 versus 2.0, bidirectional or unidirectional streaming, current voices and languages, prompt/prosody controls, subtitle timing validation, billing and concurrency, authorized replicated voices, privacy, error handling, and output QA. Do not use for mainland-China Volcengine Doubao Speech endpoints.
Create dark cinematic SaaS/product motion videos in the Presenton-like "magic UI" style: black starfield stage, bottom purple glow, large kinetic type, cyan-to-magenta CTA buttons, prompt cards, floating UI templates, integration badges, model rings, export objects, speed-blur transitions, and short product capability arcs. Use when the user asks for "这种风格", "同款风格", dark SaaS promo, AI product demo, magic UI product video, or a style-level creation inspired by a reference video. Do not use for pixel-level or frame-accurate replication; use reference-video-replica-qc first for that.
Definitive reference for FFmpeg and ASS/SSA animation timing units, optimal durations, and best practices. PROACTIVELY activate for: (1) Animation timing questions, (2) ASS subtitle timing, (3) Karaoke timing tags, (4) Caption duration calculation, (5) Transition duration selection, (6) Fade/zoom timing, (7) Frame rate considerations, (8) Platform-specific timing (TikTok/Shorts/Reels), (9) Readability formulas, (10) Audio-video sync tolerances. Provides: Complete time unit reference tables, optimal duration guidelines, psychology-based timing recommendations, caption readability formulas, and platform-specific timing profiles.
Complete FFmpeg video analysis and quality control filters for automation and broadcast workflows. PROACTIVELY activate for: (1) Detecting black frames (blackdetect, blackframe), (2) Finding blurry/frozen frames (blurdetect, freezedetect), (3) Auto crop detection (cropdetect), (4) Scene change detection (scdet), (5) Interlace detection (idet), (6) Quality metrics (PSNR, SSIM, VMAF), (7) Signal analysis (signalstats), (8) Frame information logging (showinfo), (9) QC automation scripts, (10) Broadcast compliance checking. Provides: Detection filters, quality metrics, analysis commands, automation patterns.
Generate and edit images with OpenAI GPT Image 2 (ChatGPT Images 2.0) on RunComfy. Documents GPT Image 2's strengths (embedded text, logos, multilingual typography, instruction precision), its 3 fixed sizes, edit-with-preservation language, and when to route to a sibling (Flux 2 / Nano Banana Pro / Seedream) instead. Calls `runcomfy run openai/gpt-image-2/text-to-image` or `/edit` through the local RunComfy CLI. Triggers on "gpt image 2", "gpt-image-2", "ChatGPT Images 2", "image 2", or any explicit ask to generate or edit with this model.
Manage your self-hosted Immich photo library through conversation: natural language and OCR search, geographic album curation, duplicate detection, people and faces, metadata repair, video frames and PDF photobooks. 94 MCP tools and 13 skills, tested live on Immich 2.x and 3.x. Install: claude plugin install immich-photo-manager
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.