Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Use Meitu's automatic video removal capability to clean subtitles from videos and restore the footage covered by subtitles. Use when the user requests subtitle removal from a video, cleaning text from footage, or removing burned-in subtitles.
Use Meitu's automatic video removal capability to clean watermarks from videos and restore the footage they obscure. Use when the user requests watermark removal from a video, cleaning footage obscured by watermarks, or improving visual cleanliness.
Build taste through volume of output, deliberate failure, and speed. Taste cannot be developed by watching. It's sharpened by making, failing, and making again. Use when designing practice routines, breaking through creative blocks, or helping someone move from analysis to execution.
World-class level design expertise - spatial storytelling, player flow, blockout methodology, and the invisible hand that guides players without them knowingUse when "level design, blockout, graybox, whitebox, player flow, level layout, combat arena, exploration space, map design, multiplayer map, pacing, weenie, landmark, gating, metroidvania, open world design, linear level, environmental storytelling, teaching through design, playtesting, heatmap, metrics, sightlines, level-design, game-development, spatial-design, blockout, graybox, player-flow, pacing, environmental-storytelling, combat-design, multiplayer-maps, open-world, linear-design, metroidvania" mentioned.
Expert knowledge for real-time and baked lighting in games - from cinematography fundamentals to engine-specific optimization, covering GI, time-of-day, volumetrics, and platform-aware lighting pipelinesUse when "lighting, light design, baked lighting, realtime lighting, lightmap, lightmapping, light probe, reflection probe, global illumination, GI, ambient occlusion, shadow, shadow cascade, time of day, day night cycle, volumetric fog, volumetric lighting, god rays, HDR, tonemapping, bloom, exposure, emissive, area light, Lumen, Enlighten, ray tracing, RTGI, light baking, lighting, global-illumination, lightmapping, shadows, GI, light-probes, reflection-probes, HDR, tonemapping, volumetric, fog, time-of-day, baked, realtime, mixed-lighting, lumen, enlighten, radiosity, ray-tracing, RTGI" mentioned.
Analyzes one or more sample articles to extract a writing style profile (文风画像), then rewrites any target article in that exact style. Trigger when the user says: "分析这篇文章的文风", "模仿这个文风改写", "克隆某人的写作风格", "用 XX 的文风改写", "style clone", "rewrite in this style", or pastes articles asking for style analysis / imitation.
Transcribe audio/video using trx CLI and post-process results with agent corrections. Use when: (1) user wants to transcribe a video or audio file, (2) user shares a YouTube/Twitter/Instagram URL for transcription, (3) user says "transcribe", "subtitles", "srt", "transcript", (4) user wants to fix/clean up a whisper transcription, (5) user asks to extract text from a video.
Unity 6 animation system guide. Use when working with Animator Controllers, animation state machines, blend trees, animation clips, Avatar system, humanoid rigs, root motion, animation events, Timeline, or Cinemachine. Based on Unity 6.3 LTS documentation.
Applies the Heath brothers' SUCCESs model (Simple, Unexpected, Concrete, Credible, Emotional, Stories) to make messages memorable and persuasive, with systematic analysis, targeted improvements, and scoring (0-18 stickiness scorecard). Use when making messages more memorable or compelling, preparing presentations, crafting pitches or campaigns, or when user mentions stickiness, making ideas stick, persuasion, SUCCESs framework, or Heath brothers.
Structured exposure habits for building taste through relentless, intentional consumption. Not passive scrolling. Active intake across design, architecture, film, music, food, and culture. Use when building a taste practice, advising on professional development, or diagnosing why someone's design range feels narrow.
Edit a raw talking-head video through transcription, silence trimming, mistake review, visual storytelling, motion graphics, and verified HyperFrames rendering. Use for a complete edit; route a single requested operation directly to its specialist skill.
Complete Docker FFmpeg deployment system. PROACTIVELY activate for: (1) Docker FFmpeg image selection (jrottenberg, linuxserver), (2) GPU passthrough (NVIDIA, Intel, AMD), (3) Volume mounting and permissions, (4) Docker Compose video processing, (5) Kubernetes FFmpeg jobs, (6) Custom Dockerfile builds, (7) Windows/Linux/macOS Docker usage, (8) Resource limits and optimization, (9) Watch folder automation, (10) Production container patterns. Provides: Image comparison tables, GPU Docker commands, Compose examples, K8s manifests, troubleshooting guides. Ensures: Consistent, isolated FFmpeg environments across platforms.
Authoritative Python-FFmpeg parameter integration reference ensuring type safety, accurate parameter mappings, and proper unit conversions. PROACTIVELY activate for: (1) ffmpeg-python library usage, (2) Python subprocess FFmpeg calls, (3) Caption/subtitle parameter mapping (drawtext, ASS), (4) Color format conversions (BGR, RGB, ABGR, ASS &HAABBGGRR), (5) Time unit conversions (seconds, centiseconds, milliseconds), (6) Type safety validation (int, float, string), (7) Coordinate systems, (8) Parameter range enforcement, (9) Frame pipe handling, (10) Error detection for type mismatches. Provides: Complete parameter type reference, color format conversion tables, time unit conversion formulas, validation patterns, working Python examples with proper typing.
Analyzes a game system against design principles — economy health metrics, difficulty curve, reward schedules, and progression cost curves. Flags imbalances with severity levels. Use after creating an MVP plan with game-architect. Extract the analysis scope (economy/difficulty/progression/rewards/all) from the user's message.
This skill should be used when the user asks to "generate an icon", "create a favicon", "make an app icon", "create iOS icon", "create Android icon", "generate PWA icons", "make desktop app icon", "create Windows icon", "create macOS icon", "app store icon", "Play Store icon", "App Store icon", or needs AI-generated icons with platform-specific sizing.
One MCP, 30+ image and video models. Use this skill when the user wants to generate images or videos via Higgsfield (Seedance 2.0, Sora 2, Veo, Kling, Nano Banana Pro, GPT Image 2, Flux, Hailuo, Soul, Cinema Studio, etc.) without juggling per-provider API keys. Triggers on "higgsfield", "higgsfield mcp", "use higgsfield", or any image/video generation request when the user already has a Higgsfield subscription.
Master router for immersive visual experiences combining React Three Fiber, shaders, particles, post-processing, GSAP animation, and audio. Use when building 3D web experiences, visualizers, creative coding projects, interactive installations, or any project requiring multiple visual technologies. Dispatches to 6 specialized domain routers covering 29 total skills.
Use this skill when the user says "develop a topic", "analyze a topic", "help me write an outline", "will this topic go viral", "help me diagnose this topic", "is this topic worth doing", or "how should I improve this topic". Even if the user merely shares a topic and asks for an opinion, proactively enter the diagnosis flow rather than giving a simple answer. Driven by cognitive hijacking theory, it has four modules: perspective collision (challenging the topic's premise), topic diagnosis (graded as 🚲🛵🚗✈️), outline design (emotional peak design), and style validation (alignment with li-writer style), generating a complete development report and saving it to a file. Do not trigger for: merely recording a topic (use li-recorder) or writing a script directly (use li-writer). Use when the user wants to "develop a topic", "analyze topic potential", "write an outline", "will this topic go viral", or needs full topic diagnosis + outline design. Trigger even if the user just shares a topic and asks for opinions.
Transcribe a ScreenKite recording's mic audio, detect silence gaps in the word-level transcript, and cut those gaps from the timeline using editTimeline action=cut. Use when the user asks to "remove silences", "auto-cut dead air", "transcript-based cut", or "clean up pauses" in a .skbundle project. Do NOT use for B-roll overlay work (see use-screenkite-advanced-b-roll). Do NOT use when no .skbundle is open.
Transform creative ideas into professional, production-ready screenplays optimized for AI video generation pipelines. Converts raw concepts into structured scene-by-scene narratives with rich visual descriptions, proper screenplay formatting, and XML-tagged output for seamless integration with image/video generation tools (imagine, arch-v). USE WHEN: Converting story ideas into screenplay format, preparing content for AI video pipelines, structuring narratives for 5-10 minute short films, generating visual-rich scene descriptions for image generation. WORKFLOW: Raw idea → Scene breakdown → Visual enhancement → Professional formatting → XML-tagged markdown output OUTPUT: Markdown document with XML-wrapped scenes, rich visual descriptions, proper screenplay elements (sluglines, action, dialogue), and metadata for pipeline processing.
Sets up a Mac or Windows PC for ButterCut. Installs all required dependencies (Homebrew on a Mac, Ruby, Python, FFmpeg, WhisperX). Use when user says "install buttercut", "set up my mac", "set up my pc", "get started", "first time setup", "install dependencies" or "check my installation".
Product demo and release-notes videos with no screen recording. Rebuild the app's screens as static React from its source code, then drive them in Remotion with a scripted cursor and camera (lib/components Stage + Cursor). Use when making a product walkthrough, feature demo, release-notes or changelog video for an app whose source you can read, or when a Playwright recording would be flaky, need test data, or go stale. Triggers include synthetic demo, rebuild the UI, release video, feature walkthrough, product demo from source, scripted cursor, camera zoom on UI, no screen recording.
Use when the user wants to capture a writer's voice from a few sample articles and rewrite text to sound like that author — "蒸馏文风", "提炼XX的风格", "把这篇改成XX的味儿", "学某某笔法改写", "mimic this writer", "match this author's style", "style transfer". Works for any author, 王建硕 included.
Detect and fix AI-generated writing patterns (slop). Comprehensive detection with 50+ patterns across 3 severity tiers, structure weighted above lexical tells, scoring system, and editor mode that directly fixes problems. Use when scanning any content for AI tells, auditing drafts before publishing, checking if writing "sounds like AI", humanizing AI-generated text, or verifying content authenticity. Trigger on "check for slop", "does this sound like AI", "humanize this", "AI audit", "slop check", "clean up AI writing", or any request to detect/remove artificial-sounding patterns. Also use proactively before publishing any AI-assisted content.
Use when the user wants to improve a specific article under a fixed viral-scoring rubric, iterate on one article to raise its score, or run single-article scoring plus rewrite loops. Best for content optimization experiments where the scoring prompt stays fixed and the article text is the thing being changed.
Create visual concepts, image-generation prompts, and optional AI-generated images for product hero images, marketing creatives, social posts, ads, ecommerce PDP visuals, and general visual design tasks. Use when the user asks for visual strategy, image prompt writing, product/marketing image direction, or direct text-to-image generation. In Codex environments use the built-in imagegen tool; elsewhere use the bundled script backed by the apiz CLI.
Complete Modal.com FFmpeg deployment system for serverless video processing. PROACTIVELY activate for: (1) Modal.com FFmpeg container setup, (2) GPU-accelerated video encoding on Modal (NVIDIA, NVENC), (3) Parallel video processing with Modal map/starmap, (4) Volume mounting for large video files, (5) CPU vs GPU container cost optimization, (6) apt_install/pip_install for FFmpeg, (7) Python subprocess FFmpeg patterns, (8) Batch video transcoding at scale, (9) Modal pricing for video workloads, (10) Audio/video processing with Whisper. Provides: Image configuration examples, GPU container patterns, parallel processing code, volume usage, cost comparisons, production-ready FFmpeg deployments. Ensures: Efficient, scalable video processing on Modal serverless infrastructure.
Platform-oriented writing assistant for long-form serialized web fiction. Handles new project setup, chapter continuation, single-chapter revision, project memory maintenance, long-work volume governance, and chapter quality checks. Built-in three-layer writing principles (chapter/scene/language) fit mainstream genres such as wish-fulfillment, romance, and power struggles, supporting the whole process from premise to ultra-long works. For more information follow 抖音君黎.
Synthesize and render MIDI/MusicXML files to audio (WAV/MP3) and generate Strudel live-coding patterns for creative music manipulation. Uses analysis data (tempo, key, chords, dynamics) from audio-to-midi for intelligent ghost-note removal and accurate reproduction. Two workflows: (1) Convert MIDI/MusicXML to Strudel/TidalCycles code with note durations for interactive tweaking in the browser at strudel.cc, (2) Render MIDI/MusicXML to WAV/MP3 with FluidSynth or built-in synthesis, with controls for tempo, transposition, instrument override, and ghost-note cleaning. Trigger when the user wants to: play a MIDI file, render MIDI to audio, synthesize MIDI, convert MIDI to Strudel code, create live-coding patterns from MIDI, change instruments, transpose or tempo-shift, clean ghost notes, or any MIDI playback/synthesis task.
Record browser or Web UI interaction demos as optimized GIFs using the available browser-control workflow, optional Playwright Videos for higher capture frame rates, and deterministic encoding, then attach the GIF to a pull request with `gh --attach`, falling back to a dedicated assets branch where attach cannot apply. Use when asked to make, record, or generate a GIF that demonstrates a browser workflow, and for every pull request that changes product-user-visible GUI behavior, which MUST include a GIF recorded from the pull request's real server and model flow.
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.