Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Create professional infographics using Nano Banana Pro AI with smart iterative refinement. Uses Gemini 3 Pro for quality review. Integrates research-lookup and web search for accurate data. Supports 10 infographic types, 8 industry styles, and colorblind-safe palettes.
Use this skill when the user says "generate a script", "write a video", "write a long article", "generate a short video script", or "help me write this episode". Based on the practitioner equal-dialogue style (authenticity 9/10, actionability 9/10), first run a content pre-check (goal + cognitive gap), then generate the complete script (with visual annotations) and save it in standard md format, and finally route to li-opening or li-cover. Should not trigger: optimizing only the opening (use li-opening), generating only a cover (use li-cover), or topic analysis (use li-topic). Use when the user wants to "generate script", "write video", "write article", "write long-form content", or needs a full video script or article in the practitioner equal-dialogue style.
Kinetic typography, logo animations, and stylized motion design skill. **Triggers when:** - User wants text animations or kinetic typography - Request involves logo reveals or title sequences - Content focuses on visual impact over education - User mentions "title", "intro", "logo", "text animation", "kinetic" **Capabilities:** - Kinetic typography with dynamic text effects - Logo reveals and brand animations - Particle effects and visual flourishes - Modern motion design patterns - Attention-grabbing title sequences
Provides Needle Engine context for web-based 3D projects built on Three.js with the @needle-tools/engine component system. Use this skill whenever the user is working with Needle Engine components, GLB files exported from Unity or Blender, Vite configs with needlePlugins, TypeScript classes extending Behaviour, or anything involving @needle-tools/engine imports. Also trigger when the user mentions "needle engine", "needle tools", serializable decorators (@serializable, @syncField, @registerType), the <needle-engine> web component, 3D web apps using a component system on Three.js, or 3D scenes loaded from GLB in a web context — even if they don't explicitly name the engine. Trigger even when the user is just reporting a bug or error without mentioning Needle Engine — check if @needle-tools/engine is in package.json or imports. If the project uses Needle Engine, always load this skill.
Script-writer that drafts presentations, essays, emails, and slides using only the cognitive and persuasive heuristics from Patrick Winston's "How to Speak" lecture, rejecting conventional writing advice in favor of Winston's evidence-based rules.
Design production-ready prompts, shot plans, storyboards, extensions, and video edits for Jimeng Seedance 2.5. Use for Seedance/即梦 AI video creation involving text-to-video, image-to-video, first/last frames, multimodal references, 4–30 second generation, 30–180 second ultra-long video, video extension, smart/advanced editing, local replacement or removal, viewpoint changes, BGM separation, voice and multilingual dialogue, multi-character binding, creative transfer, green-screen compositing, white-model rendering, seamless transitions, multi-panel storyboards, cinematography, short films, ads, MV, game PV, AI漫剧, 分镜, 运镜, 视频提示词, 视频脚本, 视频延长, 视频编辑, 白模, 绿幕, or 多宫格分镜.
Plan direction and editing for an already analyzed video, then write and validate timestamped Chinese narration; also handle promotional titles, on-screen text revisions, and copy updates for existing short videos. Standard planning takes work_dir's agent_narration_brief.md and vlm_analysis.json as input; copy revisions take the current edit's project and content evidence. Planning outputs recap_story_plan.json, visual_audio_board.json, optional style_card.json, clip_plan.json for cut mode, and validated narration.json; promotional-copy-only tasks deliver proposals or update an existing packaging plan. External data note: only the advisory review.py connects to the internet. It sends the full narration script, understanding evidence, and text excerpts from planning files to the MiMo chat interface (MIMO_API_KEY / MIMO_API_URL; no video, images, or audio are sent). When used alone, it runs only when explicitly executed; end-to-end orchestration runs it once before TTS by default. Disable it with --no-review-narration / REVIEW_NARRATION=0 (except when strict review is enabled); validate.py and lint run locally only. Triggers: narration, write narration, video voice-over, promotional title, on-screen text revision, copy update, narration script, write script, narration copy, editing ideas, direction ideas.
Analyze videos into a structured understanding index: scene detection, ASR transcription, per-scene VLM observations, silence windows, a fused timeline, and a writing brief. Use to understand, index, or summarize videos, and as an analysis stage before further creative work. Input a video file; outputs scenes.json, asr_result.json, vlm_analysis.json, silence_periods.json, timeline_fusion.json, and agent_narration_brief.md. Triggers: video understanding, analyze video, index video, understand video.
Apply an evidence-derived 3D cartoon character short workflow observed across published GoodCase examples by Zarnab_with_Ai. Use when planning, storyboarding, or refining an AI video prompt in this motion pattern, including requests for 视频提示词, 分镜, 动作复刻, reference images, source video, or special generation steps. This is an unofficial synthesis and must preserve source attribution.
The ONE tool for UGC video. Give a `script` (any length) and optionally a `person` description, an `image` (photo), or a `character` (saved char_… or sheet URL); it returns the finished vertical video. Short script → one clip; long monologue → full multi-take (never trimmed); pass `broll_url` → narrated b-roll overlay. Captions are OPT-IN — ASK the user if they want them (and which style) before generating; set `captions:true` only if they say yes. You never pick a sub-tool.
Humanizes AI-generated text with a 5-pass rewrite workflow in English and six other languages, optional saved preferences, and optional stylometric voice matching from a writing sample. Works without Slop or Not. When Slop or Not Pro is reachable, adds on-device AI detector scoring (English only), native-language readability checks, Text Cleanup, and cleanup stats. Use when the user invokes /agentic-humanizer or asks to humanize text.
Run registered ComfyUI workflows through the fast comfyui-skill CLI, and use the official local Comfy MCP for live template, node, model, validation, and orchestration capabilities. **Use this Skill when:** (1) The user requests to "generate an image", "draw a picture", or "execute a ComfyUI workflow". (2) The user has specific stylistic, character, or scene requirements for image generation. (3) The user asks you to import, register, sync, or configure saved ComfyUI workflows for later reuse.
The whole setup in one command — run the entire reel factory end to end. Claude takes a proven viral format, studies it frame by frame, writes an original hook + script in your voice, builds the finished reel (on-screen text, scene per beat, voiceover/music/captions/animations), writes the caption + hashtags, and wires the comment→DM lead loop — then repeats it 3× a day. This is the orchestrator that runs the other skills in order. Use when the user says "run the content factory", "make me a reel", "set up my whole content system", "how do you make these reels", "do the whole thing", "build my content machine", or "post for me every day".
This skill MUST be used for growth-oriented article, video, newsletter, and social content work, including content analysis, ideation, rewriting, packaging, production, repurposing, and repeatable editorial workflows. MUST trigger on natural-language requests such as 幫影片想標題、拆解成功內容、 規劃內容主題、把文章改成社群貼文、設計內容生產流程; users need not name the skill or explicitly say growth. SHOULD be used when the context calls for improving content reach, retention, engagement, or conversion. MUST NOT trigger solely because a request contains a URL, notes, or general proofreading without a content-growth purpose. Supports source-led and brief-led work.
The psychology behind short-form content that holds attention. Use it whenever you plan, write, review or edit a short-form video script (Reels, TikTok, Shorts), a hook, or a social post. It teaches how a viewer's mind works, never a template to fill in, so apply the principles silently instead of turning them into a checklist or format.
Text-to-speech conversion using node-edge-tts npm package for generating audio from text. Supports multiple voices, languages, speed adjustment, pitch control, and subtitle generation. Use when: (1) User requests audio/voice output with the "tts" trigger or keyword. (2) Content needs to be spoken rather than read (multitasking, accessibility, driving, cooking). (3) User wants a specific voice, speed, pitch, or format for TTS output.
Provides editorial judgment, with sourced figures, on cuts and pacing, J- and L-cuts, jump cuts, b-roll, hooks and retention, captions, on-screen text and graphics, zooms and punch-ins, audio levels in LUFS, music ducking, sound effects, basic color, thumbnails, titles, chapters and licensing. Use when planning, scripting, editing, reviewing or publishing any video for YouTube, Shorts, Reels or TikTok: a talking head, vlog, podcast or interview, screen tutorial, explainer, vertical clip from a long video, or when looking for royalty-free sound effects, music, animated graphics or a meme to recreate. Works with any tool (Premiere Pro, DaVinci Resolve, Final Cut Pro, CapCut, ffmpeg, Remotion, HyperFrames). Includes recipes by video type, a review checklist, a CC0 sound library, a CC0 graphics kit for HyperFrames, a meme reference with Content ID risk, and tested templates for narrated screen tutorials.
Turn original SRT and design into editable B-roll through an independent Director, focused chapter creators, and visual review. Use for SRT-to-video production or revisions; preserve existing legacy projects and explicit Remotion choices.
Complete fal.ai optimization system. PROACTIVELY activate for: (1) Queue vs run performance, (2) Parallel request batching, (3) Streaming for real-time UI, (4) WebSocket for interactive apps, (5) Model cost comparison, (6) Image size optimization, (7) Inference step tuning, (8) Webhook vs polling, (9) Result caching by seed, (10) Serverless scaling config. Provides: Parallel patterns, cost strategies, caching examples, monitoring setup. Ensures optimal performance and cost-effective usage.
AI screenplay/script writer for film and video. Use when: writing scripts, screenplays, dialogue, scenes, story beats, character development. Generates production-ready scripts. Keywords: script, screenplay, writing, dialogue, scenes.
Fotor AI image generator and AI video generator for photo editing, background remover, background replacement, product photos, ad creatives, social media graphics, poster and banner design, image upscaling, photo restoration, portrait enhancement, text-to-video, and image-to-video. Built for e-commerce, marketing, branding, and content creation.
Use when creating, choosing, changing, or integrating game art, animation, or audio. Research comparable assets, establish a consistent direction, use suitable production tools, and inspect the result at gameplay scale and in motion. Applies to sprites, textures, models, effects, music, sound, and voice.
Edit images on RunComfy — this skill is a smart router that matches the user's intent to the right edit model in the RunComfy catalog. Picks Nano Banana Edit (batch up to 20, identity-preserving default), OpenAI GPT Image 2 Edit (multilingual in-image text rewrite, multi-ref composition, layout precision), Flux Kontext Pro (single-ref high-fidelity local edit), or Z-Image Turbo Inpaint (mask-driven precise region edit). Bundles each model's documented prompting patterns so the skill gets sharper edits without burning iterations on the wrong model. Calls `runcomfy run <vendor>/<model>/edit` through the local RunComfy CLI. Triggers on "image edit", "edit image", "image-to-image", "i2i", "swap background", "remove object", "rewrite headline", or any explicit ask to edit a single or batch of images
Create premium hand-drawn architecture and process diagrams in the Lanshu animated GIF style, with editable .excalidraw files, static PNG previews, and genuinely animated GIFs with moving flow highlights. Use this skill whenever the user asks for 岚叔动态架构图, Excalidraw-like diagrams, DailyDoseOfDS-style black-background sketches, animated architecture/process GIFs, polished flowcharts, visual explanations of articles or system designs, or asks to replicate or improve a reference diagram with hand-drawn animated effects.
Dual-renderer animation orchestrator for METHOD_LAB/historical episodes. By default, CODEX_DIRECT production does not use this skill; use it only when the user explicitly requests the legacy AST, a Remotion/HyperFrames comparison, or restoring a historical episode.
Use when user needs to convert audio or voice recordings to text. Provides necessary command usage instructions for converting audio to text. Supports multiple languages. Triggers: 语音转文字, 音频转文字, 语音识别, speech to text, STT, 转录, 听写, 会议记录转文字, 录音转文字, transcribe, 音频识别, voice to text, 语音转换
Generate speech, music, and sound effects using ModelsLab's v7 Voice API. Supports text-to-speech, speech-to-text, speech-to-speech, music generation, sound effects, dubbing, song extension, and song inpainting via ElevenLabs and Inworld models.
Write like a human, not a language model. Avoid AI-tell vocabulary, formulaic structures, and hollow emphasis. Apply to ALL written output including prose, documentation, comments, and communication. Use when drafting prose, documentation, comments, or any written output that should sound human.
This skill should be used when the user asks to "add a glow", "blur the background", "add a screen effect", "mask a sprite", "add post-processing", "shader effect", "add lighting", "flashlight cone", "vision cone", "normal maps", "make the screen flash", "color grading", "tint a sprite", "screen shake with effects", "damage flash", "hit flash", "vignette", "bloom", "convert my preFX/postFX code", "BitmapMask replacement", "custom shader", "SpriteGPULayer", "Mesh2D", "stencil mask", or any Phaser 4 visual effect, filter, mask, light or renderer question.
Unity 6 audio system guide. Use when working with sound effects, music, AudioSource, AudioClip, AudioListener, Audio Mixer, spatial audio, or audio optimization. Covers 3D sound, mixer groups, snapshots, and audio import settings. Based on Unity 6.3 LTS documentation.
Use when the user wants a Vox-style / editorial motion-graphics look — the "printed, not lit" register: cream-paper canvas, halftone dots, heavy ink type, a single yellow highlighter marker, kinetic captions, hand-drawn annotations, and paper slap-ins. Trigger on "Vox style", "Vox-style explainer", "editorial motion graphics", "editorial collage animation", "kinetic typography explainer", "halftone / paper / newsprint animation", "highlighter / marker text animation", "animated map or data reveal in the Vox look", or "make it look like a Vox video". Also use when someone wants to CAPTURE a style they saw somewhere into their own reusable pack — see references/capture-your-own-style.md. Builds on HyperFrames (renders video from HTML); install heygen-com/hyperframes first.
Use when working with the Adobe I/O CLI plugin @adobe/aio-cli-plugin-ffcpe (FFCPE — Firefly Creative Production for Enterprise). Covers installation, authentication, env overrides, and all `aio ffcpe catalog` commands. Use for terminal workflows, scripting catalog changes, or explaining how developers should invoke this plugin. Some users still say **Workflow Builder**; that name referred to the same product area — the CLI topic is **`ffcpe`** (`aio ffcpe`).
A manuscript editor that helps surface the truth in people’s writing and cut the filler. Focuses on tightening structure and removing fluff: cuts formulaic openings, weak intensifiers, false parallelism, and slogan-like endings. Offers a quick-cut mode (delete only, no rewriting) and a full mode. Accepts any draft and can use a voice-dissolver voice list as a protected zone. Triggers: “help me edit this,” “help me revise this,” “it doesn’t read well but I don’t know why,” “too much fluff,” “it doesn’t sound like me,” “polish this,” “make it flow,” “make it more concise,” “cut the filler,” “make it sound more human,” “it’s too wordy.” Not for: heavy AI tone when the structure is fine (→ plainspoken-writing); content review (→ chinese-write-checker); title, layout, or punctuation polishing (→ packaging-workshop).
Eva Teaching 2.5.0 provides beginner tutorials, benchmark teardowns and study, and manuscript review and editing. Use for /eva-teaching, /eva-new-user, /eva-benchmark-copy, teaching me how to use Eva, tearing down third-party self-media content, or submitting your own draft for review and explanation, teaching me how to revise. Benchmarks default to fast direct output, expand when depth is specified, and invite judgment only when hands-on practice is specified; manuscripts default to explaining problems and editing as authorized. Direct creation, standalone openings or pacing optimization, AI-flavor special topics, pre-publish final checks, published performance reviews, account positioning, and guided learning from materials still use the original entry points.
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.