Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
AI image generation and editing (都爆鸭) — use it when one sentence should yield one image: text-to-image, image-to-image, image editing, key visuals, or a single illustration. Give a description and it produces an image; give a reference image (a local file works too) and it edits on that basis. Slow operation: usually 1–2 minutes, up to 4 minutes; don't interrupt or retry while waiting (a retry means paying twice for the same image). After generation it saves a local file and gives you the path. Triggers: 画张图, 帮我画, 生成图片, 生成一张图, AI 出图, 出图, 文生图, 图生图, 改图, 改一下这张图, P 一下, 配图, 配张图, 配一张插图, 来张主视觉, 做张视觉图, 按这个描述画, 封面, 封面图, 公众号封面, 做张封面, 配张封面图, 首图灵感. Not for: wanting only cover formulas and reference data (no finished image) — use dby-api; placing images into an article and saving a draft — use dby-publish.
Cover design for WeChat official accounts and Xiaohongshu. Fixed 3:4 vertical format, 10 composition styles to choose from, real person on camera, generates prompts directly usable with image generation models. Supports multiple reference images: image 1 is fixed as the face, image 2 and onward are assets such as UI screenshots and product images. Keywords: cover design, prompt generation, WeChat official account cover, Xiaohongshu cover.
Add "human feel / loss of control" to Chinese web fiction and cure deep AI-ness. Use when users complain that a draft "reads like AI / like a manual / too neat and smooth / emotions too even / characters too normal / no human feel / too neutral", or want the writing to feel more like a real person wrote it (not polishing rhetoric, but adding a sense of losing control). Use irrational behavior, incomplete emotional expression, logic that does not fully close, and personal bias to turn a "manual" into an "emotional record".
Humanize and de-slop AI-sounding prose while preserving meaning, specific detail, and genuine human quirks. When a voice sample or style profile is available, rewrite in that writer's actual voice. Remove recurring AI tells such as inflated significance, stock vocabulary, uniform rhythm, excessive hedging, formulaic structure, and formatting artifacts. Also clean suspicious invisible Unicode and copy-paste residue conservatively, without treating them as proof of authorship. Use whenever a user wants to humanize, de-AI, de-slop, or de-robotify text; fix prose that sounds corporate, generic, like ChatGPT, or otherwise off; match a draft to a writer or named author; edit for authentic voice and rhythm; or clean zero-width characters, exotic spaces, bidi controls, and similar hidden text residue.
Edit images with precision using ImageMagick 7, sips, transparent-background, and Pillow — with intelligent routing to AI semantic editing (Gemini/nano-banana-pro) for content-aware operations. Covers cropping, resizing, compositing, annotating, format conversion, color adjustment, background removal, batch processing, and montage creation. Triggers on image editing, crop, resize, composite, annotate, remove background, format conversion.
Mathematical visualization skill for equations, proofs, and geometric concepts. **Triggers when:** - User mentions equations, formulas, or mathematical expressions - Request involves mathematical proofs or derivations - Content includes geometric relationships - User mentions LaTeX, calculus, algebra, geometry, trigonometry - Patterns: "equation", "formula", "prove", "derive", "graph", "plot" **Capabilities:** - LaTeX equation rendering with color-coded components - Function graphing and transformations - Geometric constructions and proofs - 3D mathematical surfaces - Step-by-step derivations with highlights
USE FOR finding images on the web by a text query or a reference image (one input per request), powered by Octen. Returns ranked images with thumbnail, source page, dimensions, and description. In Beta. Contact us to request beta access at https://octen.ai. Use it for general image search (photos, diagrams, products, screenshots). For UI design references with structured style tokens and HTML/CSS snippets, use the octen-design skill instead.
This skill should be used when converting academic papers into promotional and presentation formats including interactive websites (Paper2Web), presentation videos (Paper2Video), and conference posters (Paper2Poster). Use for paper dissemination, conference preparation, creating explorable academic homepages, generating video abstracts, or producing print-ready posters from LaTeX or PDF sources.
Use when writing a social post on a topic or a reply to someone's post — X/Twitter, LinkedIn, Bluesky, Threads — and the obvious take is not good enough. Triggers include "reply to this post", "write a post about", "quote this", "tweet about", "my replies all sound the same", "what's a non-obvious take on", "build in public post", a pasted post with "what would you say to this", or a post URL. Do NOT use for analytical work like debugging, code review, or implementation tasks.
Generate, convert, preview, and integrate retro game sound effects with raysan5/rfxgen. Use when a user needs coin, laser, explosion, power-up, hit, jump, or blip SFX; wants `.rfx` parameters converted to WAV/raw/C headers; needs batch CLI generation; or must build and troubleshoot rFXGen. Triggers on: rFXGen, rfxgen, sfxr, chiptune SFX, .rfx, procedural game sound, or sound preset.
Transform a user-supplied photograph into an original vertical 3:4 art-directed poster using watercolor, pop-art screenprint, expressive painting, ink-wash, relief print, editorial surrealism, or a controlled hybrid. Use when the user asks to reinterpret a real scene, portrait, object, vehicle, landscape, architecture, or travel photo as creative visual art, an exhibition poster, editorial poster, key visual, or unified poster series with source-specific typography, editorial micro-copy, image-title interlocking, and selected identity preservation.
Generates illustrated graphic novel narratives about scientists, mathematicians, and historical figures for intelligent textbooks, with image prompts for each panel. Use when adding historical-figure or case-study stories to a textbook's Stories section.
Plan presenter-led and UGC-style videos: choose the presenter archetype, define styling, setting, lighting, and framing, write the hook/body/CTA script skeleton, and produce keyframe prompts for the repo's FAL.ai pipeline. Use whenever the user wants a talking-head video, presenter video, UGC ad, testimonial, founder video, explainer with a person on camera, spokesperson content, or 'someone talking about my product' — even if they just say 'I need a video of a person explaining this'. Produces a Talking Head Video Spec: presenter definition, look/setting/lighting/framing decisions, timed script skeleton, and generation-ready keyframe prompts for fal-ai/nano-banana-pro.
Create, edit and evaluate SillyTavern character cards and world books (character info, worldbuilding, NPCs, scenes, events, writing style, MVU variables, EJS schemes, etc.). Covers world books embedded in character cards and standalone world books, and continues project creation and creative planning after tavern-design produces design-spec.md. Make sure to also use this skill when the user mentions keywords such as '角色设定', '人设卡', 'worldlore', 'character card', '角色卡', '酒馆世界书', 'SillyTavern', '世界书条目', '世界书配置', '蓝灯', '绿灯', '角色条目', '世界观设定', 'NPC设定', even if they do not explicitly say '世界书'. It should also trigger when the user works on creative processes such as writing basic character info, personality palettes, three-sidedness, secondary interpretation, multi-stage palettes, worldbuilding or style guides. It should also trigger when the user mentions keywords such as 'MVU', '变量系统', 'schema.ts', '变量更新', 'initvar', '变量结构', 'tavern-cards-forge'.
Implement and debug Three.js motion with AnimationMixer, clips and actions, GLTF animation playback, skeletal rigs, morph targets, cross-fades, and frame-rate-independent procedural animation. Use when playing or blending model animations, driving bones or morphs, creating timeline motion, or fixing animation lifecycle and performance issues. Triggers on: Three.js animation, AnimationMixer, AnimationAction, AnimationClip, GLTF animation, skeletal animation, bones, morph targets, crossfade, procedural motion.
Load, configure, and optimize Three.js textures: color space, wrapping, filtering, mipmaps, UV transforms, normal/AO/PBR maps, cube and HDR environments, data/canvas/video textures, render targets, texture memory, and disposal. Use when images look wrong, UVs repeat incorrectly, HDR/IBL is needed, or texture quality and GPU memory need tuning. Triggers on: Three.js texture, TextureLoader, UV mapping, color space, sRGB, mipmap, anisotropy, normal map, environment map, cubemap, HDR texture, render target, video texture.
Creates TikTok image carousels (slideshows with text overlays on photos) via the ViralBaby API. Use when the user wants to: create TikTok slideshows or carousels, find/search for background images for social media content, post or upload slideshow content to TikTok, edit slide text, or manage image collections for content creation. Do NOT use for: general TikTok account management, TikTok analytics or metrics, video editing or video creation (this is for photo slideshows only), non-TikTok social media platforms, or any task unrelated to creating visual slideshow content for TikTok.
Tired of scrubbing through video files just to get the audio you actually need? This video-to-audio-converter skill pulls clean audio tracks directly from your video files — no editing software required. Upload mp4, mov, avi, webm, or mkv files and get back high-quality audio in seconds. Perfect for podcasters repurposing recorded sessions, musicians extracting stems, journalists pulling interview audio, or anyone who needs the sound without the screen. Fast, straightforward, and built for real workflows.
Plan, script and review short-form videos (Instagram Reels, TikTok, YouTube Shorts) so they actually get watched, with hooks that do something extraordinary in the first 3 seconds. Use when the user wants a reel idea, a hook, a script, a storyboard, or wants to know why a post flopped. Built from the real winners and losers of 262 posts and 3.5M views on @jayantcreates.ai.
Visual IP Illustrations generates selectable visual-IP article-body illustrations for articles, posts, blogs, Notion docs, workflow docs, methods, processes, structures, states, metaphors, opinions, shot lists, title-removal edits, and image repair. The omitted visual IP default is Xiaohei, with localized route aliases preserved in references/routing.md; Littlebox, Tom, Ferris, Seal, OpenClaw, Go Gopher, Cai Xukun, Hermes Agent, and Linux Mascot are explicit selectable routes.
Turn a local video into metadata, transcript, and an AI-composed card-based video where the agent freely designs and writes HTML cards in conversation. Use when the user asks for VTake, vtake-cut, video takeaways, transcript cleanup, or AI-composed video repurposing.
Beginner writing dual-mode Skill. Aimed at students and users, it does not require a personal profile, accumulated long-term memory, or representative works beforehand. When users want to "first write a very strong article" but do not have enough material, experience, or style assets of their own, prefer this skill. When users explicitly say they want to "write like a certain built-in author", especially the current built-in authors Dan Koe, 粥左罗, Justin Welsh, 刘润, 梁宁, 薛辉, this skill should also be used. Even if users do not directly say "strong article mode" or "high-mimicry mode", as long as the intent is "low-barrier drafting" or "first experience an author's direction", this skill should be triggered. Do not use it in scenarios where "users already have their own mature style assets and want to write like themselves"; that is the responsibility of `writing-clone-profile`. The currently implemented scope includes: common standards, the first batch of strong article prototypes, author profile drafts for Dan Koe / 粥左罗 / Justin Welsh / 刘润 / 梁宁 / 薛辉, and the runtime convergence protocol. Hard constraint: this skill does not use form-normalized-anchors.
The ai-caption-generator skill transcribes and burns accurate, styled captions into your videos using deep speech recognition and NLP alignment. It handles mp4, mov, avi, webm, and mkv formats out of the box, making it versatile for creators, marketers, educators, and accessibility teams. Key features include multi-language detection, customizable font and position settings, speaker-aware captioning, and SRT/VTT export alongside burned-in subtitle renders. On first use, the skill auto-configures credentials via NemoVideo API, so there is zero manual setup required before your first caption job runs. Supports mp4, mov, avi, webm, and mkv.
The orchestration layer for AI-native creative production. This skill coordinates multiple AI tools—video, image, audio, digital humans, effects—into cohesive campaigns, productions, and creative systems. As AI tools proliferate, the challenge shifts from "can we create this?" to "how do we orchestrate these capabilities into something coherent?" The AI Creative Director thinks in systems, not tools. In pipelines, not one-offs. In brand consistency across AI-generated assets. This is where creative vision meets technical orchestration. The AI Creative Director doesn't just use AI tools—they compose them into creative instruments that produce at scales and speeds previously impossible. Use when "AI creative director, orchestrate AI, AI campaign, multi-tool, AI workflow, AI pipeline, coordinate AI, AI production, AI creative system, full AI production, AI at scale, orchestration, creative-direction, ai-production, workflow, pipeline, multi-tool, scale, quality-control" mentioned.
AI Movie Maker — Create Short Films and Story Videos with AI Chat. Short film ideas stuck in your head with nowhere to go? AI Movie Maker takes your story concept and builds a complete short video — scenes, transitions, music, pacing — through conversation. Describe your narrative: 'a 90-second motivational story about a runner training for a marathon' or 'a thriller intro with three tense scenes.' The AI generates scene-by-scene footage, selects atmospheric music, adds dramatic timing, and assembles the final cut. Handles multi-scene compositions, camera angle variety, emotional pacing, and genre-specific visual styles. Useful for content creators building narrative series, educators making story-based lessons, brand storytellers, and anyone who needs video narrative without a film crew. No scripting software, no editing timeline. Describe the story — receive the film. Export as MP4. Supports mp4, mov, avi, webm, mkv. Works by connecting to the NemoVideo AI backend at mega-api-prod.nemovideo.ai. Supports MP4, MOV, AVI, WebM. Free trial available.
Tell me what you need and I'll compose original, royalty-free music tailored to your exact vibe, tempo, and style. This ai-music-generator-free skill lets you describe the mood, genre, or scene — and get a custom track back without spending a cent. Perfect for content creators, indie filmmakers, podcasters, and social media producers who need fresh audio without licensing headaches. Supports mp4/mov/avi/webm/mkv video uploads so you can match music directly to your footage.
Generate a structured short-video shooting script from a topic. Emits a strict, machine-parseable shot list (3 shots by default) with image prompt + video prompt + voiceover + on-screen text per shot. Trigger when the user asks for a video script, 分镜, 短视频文案, AI视频, 短剧脚本, or wants visual prompts ready for image/video generation.
Use when designing custom voices with Alibaba Cloud Model Studio CosyVoice customization models, especially cosyvoice-v3.5-plus or cosyvoice-v3.5-flash, from a voice prompt plus preview text before using the returned voice_id in TTS.
Remove backgrounds from images using AI. Triggers include: "remove background", "transparent background", "cut out", "isolate subject", "remove bg", "make transparent", "extract subject", "background removal". Creates PNG or WebP images with transparent backgrounds.
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.