Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Plan short-form video, vlog, travel, lifestyle, documentary, or cinematic social video scripts by negotiating duration, choosing a story arc and visual style, then splitting the film into timed scenes of 4–15 seconds. Use this skill whenever the user asks for a video plan, short video plan, scene breakdown, montage plan, travel video script, TikTok/Reels/Shorts storyboard, or cinematic short-form narrative, even if they only describe a loose theme or destination
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
Use BaoCut to process video and audio: transcribe, subtitle, translate, dub, trim talking-head footage, cut short videos, add chapters, summarize and draft scripts, download from links, compress and transcode, export finished videos, or create a video from scratch. Use with media files, video links, subtitle files, or BaoCut videos. Do not use for translating plain text, writing or image creation unrelated to video, using other editors or FFmpeg, or developing BaoCut source code.
Volcano Engine AI MediaKit voiceover-editing Skill: an end-to-end workflow for cutting talking-head / voiceover content. Covers environment setup, media asset management, audio processing, voiceover editing, video export, review UI, and self-iteration. You MUST invoke this Skill when the user mentions voiceover editing, cutting voiceover video, removing pauses, processing audio, exporting voiceover output, auto-editing, removing verbal slips, or video editing in this sense—or the Chinese phrases 口播剪辑, 剪口播, 剪视频, 去掉停顿, 处理音频, 导出口播, 自动剪辑, 去除口误, 视频剪辑. Also invoke when the user uploads video or audio and requests editing
A skill that uses GLM-V native grounding capabilities for coordinate conversion, bounding-box visualization, and more. GLM-V native grounding can locate any target specified by the prompt in an image and output relative coordinates normalized to 0-1000 based on image size. Coordinate formats include 2D bounding box (default), 2D points, and 3D bounding box. GLM-V also supports spatiotemporal localization and tracking of multiple prompt-specified targets in videos, outputting 2D bounding boxes per second.
Generate social media hook video prompts for TikTok, Instagram Reels, and YouTube Shorts with Seedance 2.0 (Higgsfield). Use when the user wants a scroll-stopping hook, viral short video, engaging opening, TikTok, Reels or Shorts content, or any social-media-optimized video. Trigger for social media videos, TikTok, Instagram Reels, YouTube Shorts, viral videos, hooks, scroll stoppers, short videos, trends, engagement, views, or any social media video request. Also applies to requests such as "make something that gets views" or "viral content".
Draws original cosy animal scenes as hand-authored SVG cards (480x360) in a grainy gouache, risograph-like style. Identifying traits are flat shapes with no outlines on a hot-pink card; soft 3-4 stop gradient shading on every rounded form, darker toward edges and base; a fine dark multiply grain that gathers at each shape's edges; a chalky purple-pink shadow pooled under the group; and big glossy eyes with cream star highlights, framed by grained autumn leaves and sparse cream sparkles. It covers design and coding moments with one cute animal at a laptop, monitor, desk, window, mug or book stack. Use when someone asks for grainy, gouache, riso, risograph, textured, cosy, cozy, autumn, hygge, storybook or children's-book style animal art, or for an empty state, onboarding step, 404 page, blog header, feature spot or marketing card with a warm handmade feel.
Draws original hand-written SVG spot illustrations in the One-line style, joyful action figures doing design and coding work. One continuous uniform black line (2.3 px at 480 wide) loops and doubles back to draw the whole figure. Flat coral, periwinkle and white patches sit behind the line, wobbly and deliberately knocked 2–6 px out of register like a misprinted riso. Tops carry coral stripe lines, and there is no ground plane, only a squiggle, a curl or a few speed strokes, all on warm beige paper. Covers the palette, a route-building toolkit (tube limbs, hand and shoe loops, wobble patches, masks), poses, composition and a render-critique loop. Use when asked for one-line, single-line, continuous-line, misregistered, riso-style or playful doodle illustrations, for empty states, onboarding, 404 pages, blog headers, feature spots, launch announcements or marketing cards.
Detect and humanize AI-generated Chinese text. 20+ rule detection categories + statistical features + scene-aware LR fusion (rule × 0.2 + LR × 0.8) trained on three scenes: general / academic / longform 长文本 (≥1500 字)。Unified CLI: ./humanize {detect,rewrite,academic,style,compare}. 8 style transforms (casual/zhihu/xiaohongshu/wechat/academic/literary/weibo/novel)。 Multi-paragraph rewriting + best-of-N humanize。165 replacement patterns + CiLin 同义词词林 38873。 Academic paper AIGC reduction for CNKI/VIP/Wanfang (知网/维普/万方 AIGC 检测降重)。 Pure Python, no dependencies, offline。v5.0.0 — HC3 fused 95%, 学术 -65 / 工作汇报 -83 / 长篇博客 -55。 Use when user says: "去AI味", "降AIGC", "人性化文本", "humanize chinese", "AI检测", "AIGC降重", "去除AI痕迹", "文本改写", "论文降重", "知网检测", "维普检测", "AI写作检测", "让文字更自然", "detect AI text", "humanize text", "reduce AIGC score", "make text human-like", "去ai化", "改成人话", "去机器味", "降低AI率", "过AIGC检测", "长文本改写", "小说改写"
Video understanding and transcription with intelligent multi-provider fallback. Use for: (1) transcribing video or audio; (2) understanding video content, including visual elements and scenes; (3) analyzing YouTube videos by URL; (4) extracting information from local video files; (5) getting timestamps, summaries, or answers to questions about video content. Automatically selects the best available provider based on configured API keys, preferring full video understanding (Gemini/OpenRouter) over ASR-only providers. Supports model selection per provider
Create any visual design using Adobe Express templates: flyers, posters, social media posts (Instagram, Facebook, LinkedIn), business cards, invitations, greeting cards, resumes, cover letters, brochures, newsletters, certificates, presentations, YouTube thumbnails, email headers, logos, menus, and labels. Use this skill whenever the user wants to make, design, or build any visual, even if they just say "make me a flyer", "design a poster", "I need something for Instagram", "create an event invite", or "make a business card". Also handles browsing templates, editing text, replacing images, changing backgrounds, animating, and exporting designs. Access: Signed-In required | Gen AI: off by default — image replacement only where the surface permits generative AI (e.g. Codex); none on Claude
Generate captions (descriptions) for images, videos, and documents using ZhiPu GLM-V multimodal model series. Use this skill whenever the user wants to describe, caption, summarize, or interpret the content of images, videos, or files. Supports single/multiple inputs, URLs, local paths, and base64 (images only).
Expert blueprint for horror games: sawtooth tension pacing, Director macro-AI, sensory predator AI, sanity/stress FX, and scarcity loops. Use when building psychological/survival horror, dual-brain stalker AI (cheating Director + honest LoS/sound), flashlight/fog atmosphere, or safe-room saves. Keywords: horror_game, tension_pacing, director_system, sensory_perception, sanity_system, volumetric_fog, AI_reaction_time.
Automatically generate an AI image on Higgsfield using Playwright browser automation. Use when the user has an image prompt and wants to generate it on Higgsfield Soul 2.0 or Nano Banana Pro. Triggers on requests like "generate image on higgsfield", "create image", "auto-generate image", "make the image on higgsfield", or any request to submit an image generation job. Requires Playwright MCP tools.
Generate high-quality images from text prompts using Volcano Engine Seedream models. Supports multiple artistic styles and aspect ratios. Use this skill when users want to create images from text descriptions, generate artwork in various styles, create visual content for creative projects, or need AI-powered image generation capabilities
Agent tooling for Motion Canvas — seek, screenshot, scene graph inspection, settings control, and rendering via HTTP API. Use when an agent needs to visually verify, screenshot, inspect or render a running Motion Canvas editor. Requires a browser with the editor open.
Use this skill when the user requests to generate, create, compose, or produce music or songs — background music, theme songs, jingles, or instrumental tracks. Generates a song from a style/mood prompt and optional lyrics via the MiniMax music API.
Build a risk-matched whitebox or the approved production game for its target runtime. Turn GAME_DESIGN, and ART_DIRECTION when production begins, into a minimal BUILD_BRIEF and a runnable candidate that can be iterated with replayable evidence. Use for prototype the riskiest design question, implement the approved game design, or turn this design into a running game. 游戏构建执行。先按最大风险做白盒，或在正式生产时把批准后的 GAME_DESIGN 与 ART_DIRECTION 压缩成最小 BUILD_BRIEF，在目标运行环境中实现可运行候选并用可回放证据迭代。用于验证高风险设计问题、构建游戏原型或实现批准方案。
Design game concepts from a novel. From SOURCE_BIBLE and PRODUCT_BRIEF, compare meaningful alternatives on the dimensions still unlocked, then select a playable direction using hard vetoes and explicit trade-offs. Use for what game should this novel become, compare game concepts, choose a game direction for this book, decide whether this novel should become an interactive story. 小说游戏概念设计。根据 SOURCE_BIBLE 与 PRODUCT_BRIEF 比较未锁定维度上的有效替代方案，用硬否决和关键取舍选择可玩方向。用于判断小说适合做成什么游戏、该不该做成互动叙事、比较游戏方案等需求。
Draws original design-and-coding scenes as SVG cards (480x360) in a framed-panel cartoon style. Identifying traits are a soft pastel card with a periwinkle-violet panel framed in thin navy, outline-only clouds and a pale lavender floor band with dashes inside it; L-shaped corner brackets outside two corners; white speech bubbles, UI cards and bushes that break out over the frame edge; and a semi-real character with a thin navy outline (1.25 px), muted flat fills, navy hair with violet highlights, blush and short paired hatch marks. Covers people working with laptops, screens, UI parts and devices. Use when someone asks for a framed panel, window-frame or picture-frame illustration, pastel violet cartoon, outlined flat character scene, empty-state, onboarding or 404 art, a blog header, a feature spot, a support, about or team page illustration, or a marketing card with a person.
Generates Seedream 5.0/4.0 (即梦AI) image prompts and can call the API to generate images and automatically download them to output/. Workflow: describe an idea → agent outputs a prompt for review → user confirms → agent runs generate.py. Covers text-to-image, image editing, multi-image fusion, character consistency, knowledge cards, posters, PPT backgrounds, e-commerce, avatars, and group/storyboard. Use when the user mentions seedream, 即梦, jimeng, AI生图, 文生图, 图生图, seedream prompt, 提示词, 一键生图, knowledge card, poster design, 海报, 电商图, 角色一致性, or image generation
Add b-roll motion graphics to a talking-head video. Transcribes the recording with word timings, decides line by line when the speaker is shown, when a graphic is laid over them and when a full-frame graphic covers them, draws each graphic from what is actually said (numbers, comparisons, steps, key words) in JavaScript on a canvas, and renders the finished MP4 with the original sound, plus the b-roll alone with alpha for an editor. Use when someone has a recording of a person talking to camera (a talking head, a spoken explainer, a course lesson, an ad read, a podcast clip) and wants it more engaging to watch, wants b-roll, or wants motion graphics added to it
Free YouTube video downloading, transcript extraction, and metadata retrieval using yt-dlp. Use this skill for downloading YouTube videos (single or bulk), extracting transcripts/subtitles, getting video metadata, or downloading audio-only. This is FREE (no API keys) unlike Apify. Triggers on YouTube download requests, transcript extraction, video metadata, or bulk video downloads
Turn any text, webpage, or URL into an HTML information card in the Claude/Anthropic design language, then capture it as a PNG with Playwright. Supports 14 formats: platform covers (WeChat Official Accounts, Channels, Bilibili, Douyin), image-and-text content cards (Xiaohongshu, tutorials, comparisons), social sharing cards (quotes, data, square), and long-form editorial layouts (Broadsheet, Feature, Reader, Digest). Use this skill when the user mentions “information card,” “card,” “cover,” “image-and-text post,” “layout,” “screenshot,” “generated image,” or “content card.”
Design Instagram carousel posts as self-contained HTML files in the user’s own brand. Runs a one-time brand setup on first use (name, handle, subtitle, photo, accent color, style preference). Four design styles: Bold Impact (dark premium), Tweet Post (minimal text), Clean Editorial (light/white), Product Showcase (soft gradient). Fixed 420x525px slides, Poppins font, progress bar, arrow indicators, white CTA slide, PNG + PDF export. Triggers: instagram carousel, carousel design, ig carousel, create carousel, design carousel, carousel post, make a carousel, instagram post design, carousel html, slide deck for instagram, visual carousel
This skill should be used when the user asks to "plan a scene", "scene structure", "sequel scene", "dialogue subtext", "deep POV", "psychic distance", "try fail", "scene cards", "exposition", "info dump", "flashback", "time skip", "story opening", "first page hook", "introduce a character", "scene outcome", "yes-but no-and", "chapter hook", or wants scene-level craft for drafting or revision. NOT for book-level pacing or act structure (use plot-structure), or drafting the prose of a scene or chapter (use chapter-writing).
Command-line interface for Pioneer Rekordbox 6/7 - DJ library and live-deck control via guarded SQLCipher master.db access (pyrekordbox) + virtual MIDI mapping. Provides library inspection, playlist creation/add/clear, and live-deck mixing (play/sync/crossfade/EQ). Pioneer ships no playback REST API; this harness combines the only two real surfaces (encrypted DB + MIDI) into one agent-native CLI.
This skill should be used when the user asks about "mystery", "fair play", "clue", "red herring", "romance beats", "HEA", "thriller", "ticking clock", "horror", "dread", "MG", "YA", "middle grade", "young adult", "science fiction", "sci-fi", "fantasy", "quest", "chosen one", "historical fiction", "period voice", "anachronism", "comedy", "humour", "humor", "romcom", "farce", "satire", "running gag", "serial", "episodic", "web serial", "genre conventions", or wants genre-specific structural craft for a story project. NOT for a clue check as a revision pass (use revision-continuity).
Creates original editorial illustrations where a recurring mascot character performs the idea — one caught scene by default, a hand-built explainer diagram (labeled stages, a fan-out, timeline, loop, or stack) when the structure itself is the point, or a transparent character cutout (pose-only compositing asset, no scene or text) — in one of seventeen bundled looks (sixteen print, plus a photoreal toy-brick set). Also handles "surprise me" / "random" (optionally scoped to a focus or character): rolls provenance, builds three saying candidates, picks via interactive choice or auto-pick-best (`--autopick`), and renders one image. Triggers only when the skill is directly invoked or "illo" is requested; never on generic illustrate / draw / make-an-image requests.
Draws original hand-written SVG spot scenes in the Line interior style, wide airy rooms or terraces where slim, realistic people do design and coding work. A thin uniform indigo outline (1.4 px at 480 wide) runs around mostly white fills. Sparse flat accents (teal, sage, coral, mustard, lavender, peach wood) cover under 7% of the card. Pale lavender 1 px background decor (brick clusters, pendant lamps, framed posters) sits behind, and a long floor line with end dashes runs under it all. Covers the palette, line weights, heads and hands from a parts toolkit, composition and a render-critique loop. Use when asked for line-art, thin-line or outline illustrations, indigo line art, office, desk, home-workspace, remote-work or team scenes, or interior spot illustrations for empty states, onboarding screens, 404 pages, blog headers, feature spots, landing sections or marketing cards.
Render any local PowerPoint PPTX into a fresh narrated video bundle through the public cross-platform pptx2video CLI, with Animation-Pane-authoritative object animations, unified Notes and Alt Text narration blocks, optional user script overrides, spotlight cues, block-level TTS caching, bottom-band subtitles, editable source delivery, timeline metadata, and strict QA. Use when a user invokes /pptx2video, asks for PPTX-to-video conversion, edits a PPTX and wants a rerender, needs animation-order conflict handling, or needs the Author Notes and Alt Text protocol.
Create a punchy highlight cut from a video with Adobe Quick Cut — for requests like “make a sizzle reel”, “make a highlight reel”, “quick cut this”, “cut the best parts”, “shorten this video”, “make a highlight clip”, or “summarize this video visually”. Works from the user's original request without follow-up questions and produces one highlight cut. Do NOT use it for exact-timestamp trims (“cut from 0:30 to 1:15”) or other deterministic edits — Quick Cut selects moments by relevance, not exact instructions; those need manual editing. Requires a video file (an upload or an already-referenced Creative Cloud asset).
Extracts a character list, character profiles, appearance prompts, and voice prompts from a novel or short story. The appearance prompt includes complete layout instructions for a character design sheet (left: half-body portrait + right: three full-body views + detail strip). Outputs JSON + Markdown + an interactive report.html. The report language can be specified with (--lang); defaults to Chinese and supports any language. Zero dependencies, no API key, no image generation. Use when asked to extract novel characters, analyze characters, or generate character cards or character sheets from a novel
Deconstruct a novel for game adaptation. Compress a raw novel, deconstruction library, or writing project into a SOURCE_BIBLE with cited textual evidence, extracting world rules, player verbs, spaces, character will, systems, and visual anchors—without inventing a genre yet. Use for gameable book analysis, analyzing a novel for game adaptation, or extracting game systems from a book. Gamify book deconstruction. Compress a raw novel, deconstruction library, or writing project into a SOURCE_BIBLE with necessary textual evidence, extracting world rules, player verbs, spaces, character will, systems, and visual anchors without prematurely inventing a game genre. For requests such as gamified book deconstruction or analyzing the basis for adapting a novel into a game
Create image-based PowerPoint decks by (1) turning raw article content or notes into a detailed per-slide message plan when needed, (2) turning that message plan into a slide display plan and then a visual-production plan, (3) generating one 16:9 slide image per slide with all displayed text baked into the image (English by default; multilingual slide text supported), and (4) assembling an images-only .pptx that simply concatenates those images full-screen. Use when the user wants polished, consistent visuals with extensible style packs (cinematic dark, cinematic light, cinematic editorial, illustrative cinematic, animated feature, editorial, warm pastoral, tech, youth social, academic, corporate, whiteboard sketch), prefers not to hand-layout PPT objects, or wants a repeatable prompt workflow to iterate over time.
Converts content produced by general-purpose writing Skills into a format and style suited for publication on the WeChat Official Account platform. WeChat Official Accounts favor a formal but approachable tone, polished formatting, and in-depth content
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.