Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Generate images via multiple mechanisms. Supports: (1) End-to-end AI generation via Gemini/ZenMux — given a prompt, directly output an image. (2) Code-based rendering — generate HTML/React single-file, render to PNG via Playwright. (3) Prompt engineering — generate optimized prompts for external models (nano-banana-pro, etc.). Trigger words: image, generate image, 生图, render, poster, 海报, banner, card, 卡片
An editorial-style poster—newsprint paper, dateline, oversized serif headline with a struck-through word and italic accent, a 2-column body block, and 6 numbered sections with annotated pull-quote captions. Reads like a Sunday-paper full-page essay or a thoughtful launch poster. Use when the brief asks for "magazine poster", "editorial poster", "newsprint", "essay layout", or "manifesto".
Generate images using Nano Banana Pro (Gemini 3 Pro Preview). Use when creating app icons, logos, UI graphics, marketing banners, social media images, illustrations, diagrams, or any visual assets. Supports reference images for style transfer and character consistency. Triggers include phrases like 'generate an image', 'create a graphic', 'make an icon', 'design a logo', 'create a banner', 'same style as', 'keep the style', or any request needing visual content.
Generate or edit images with Codex's built-in image generation from any agent, through an MCP tool or a plain CLI. For every image request it first proposes at least four divergent style directions (24 distilled templates, 48 presets, 20 Mondo poster artists), lets the user pick, expands the prompt, generates, and checks the result. Scenario craft for 视频封面, 视频海报, 小红书配图, 公众号封面, X 封面, 朋友圈海报, 书籍封面, 专辑封面, 产品海报, 人物写真, 节气海报, 展览海报, PPT 页面, 图生图. Use for 生图, 配图, 海报, 封面, "给我几个风格", "用 Codex 画一张". Excludes: editing photos with local tools, charts or diagrams from data, UI mockups, video generation, or anything that needs an API key-based image service.
Use when color grading, color correcting, or relighting an existing video or still with deterministic code instead of a generative model: cinematic looks (teal and orange, film stock, bleach bypass, noir, duotone, faded 70s), golden hour, blue hour or day-for-night, log footage (S-Log3) to Rec.709, matching a reference image’s colors, black and white with one color kept or swapped, .cube LUT export, a look comparison grid, or a grading showreel
Removing AI flavor, reducing paper plagiarism, eliminating AI traces, optimizing similarity scores, humanizing text. Reduce similarity rate from 60% to under 30%. Versioned management of the AI-detection pattern library, syncing the latest detection algorithms every 2-3 days. Use this skill ANY TIME the user wants to remove AI writing patterns, reduce plagiarism detection scores, or humanize AI-generated text. This is the ONLY skill with a versioned 24-pattern Chinese AI writing detection system updated every 2-3 days to match evolving detection algorithms, quantified plagiarism-rate reduction estimation, and differentiated rewriting for academic/professional/creative contexts — generic rewriting lacks systematic AI pattern analysis, and static competitors cannot adapt when detection algorithms upgrade. ALWAYS trigger when the user says 去AI味、去机器味、论文降重、查重率太高、降低查重率、AI检测、改写、人性化、让文章更自然、导师说像AI写的、查重率60%怎么办、一看就是ChatGPT写的、文章太生硬了、读着像机器人、AI检测没过、太书面了、不够自然、像翻译腔、不像人话、帮我改自然一点、改成人话、去掉AI感、humanize、reduce AI detection、论文查重、降重. Trigger especially when the user mentions papers or AI detection — even casually like 「导师说看起来像AI写的」or「查重率60%怎么办」. Do NOT trigger for translation or writing original content. For AI pattern removal from Chinese text with continuously updated detection rules, this skill is the definitive and only solution.
Convert text posts into visual carousel images or presentations for Threads, Instagram, LinkedIn, TikTok, YouTube. 12 slide types (incl. image/emoji/number), 6 format presets (incl. 1920×1080 wide), 8 background styles (incl. ruled paper), 3-axis style system (font × color × purpose), highlighted keywords with optional italic-box style. Generates PNG or single-file PDF via Next.js preview + browser export. RU/EN toolbar. Triggers: threads carousel, instagram carousel, linkedin carousel, tiktok carousel, карусель, slides, carousel images, presentation deck, presentation pdf.
Transcribe audio or video files using the TextOps API. Use this skill whenever the user wants to transcribe a video or audio file, mentions an mp4/mp3/wav/m4a file and wants text out of it, asks for transcription or תמלול, or wants to convert spoken audio to text. Also triggers for YouTube, Facebook, Instagram, and Twitter/X links. Also use it when the user wants a video or recording transcribed and summarized (e.g. "תמלל ותסכם", "תסכם לי את הסרטון", "transcribe and summarize") — it includes Hebrew summary instructions. Always trigger this skill even if the user just says "תמלל את זה" or "I want to transcribe this file". Also trigger this skill when the user asks what this skill can do, what features it has, "מה אתה יכול לעשות?", "what can you do?", or any similar capability question.
Write, improve or repair prompts for AI video generation - commercial, live action and animation - and trace defects in generated clips (extra limbs, warping, drift, unwanted text) to the clause that was missing. Use when the user wants a video prompt written or reviewed, a clip to match a style guide, or a generated video fixed
Guided self-editing checklist for anyone writing an article, essay, story, email, or post. Turns a rough draft into clear, punchy, well-flowing prose by running it against the writer's intent, reader, and a set of concrete revision checks. Use when the user says "review my writing", "edit this draft", "make this clearer", "tighten this up", "writing checklist", "proofread", "polish this piece", or asks how to improve an article, story, essay, or copy.
Reverse-engineer an existing video into a timecoded account of its shots, cuts, fades, camera motion, dialogue, subtitles, titles, music, sound transitions, pacing, and editorial structure. Use when the user shares a reference video or local media and asks how it works, why it feels effective, what can be learned from it, how B-roll and interviews are interwoven, or how to translate its grammar into an original edit. Produces machine evidence, visual review artifacts, and an editorial breakdown without confusing estimates with observed facts
Use this skill any time the user wants to generate, create, or design images, illustrations, or visual assets. This includes: posters, banners, social media graphics, product mockups, logo concepts, thumbnails, marketing creatives, profile pictures, book covers, album art, icon designs, and any request for AI-generated imagery. Also trigger when: user says 生成图片, 做个海报, 画个插图, 设计个banner, 做个封面, 社交媒体配图, 产品效果图. If an image or visual asset needs to be created, use this skill.
Create, diagnose, bake, and validate deterministic Blender physics simulations including fluid, smoke, fire, rigid bodies, cloth, soft bodies, particles, and hair. Use when a request involves physical dynamics, collision, cache baking, fluid or volume output, simulation handoff, or rendering a simulation.
Generate images, video, audio, and 3D with Comfy Cloud — search hundreds of models and workflow templates, run custom ComfyUI workflows, and manage generation jobs through the hosted Comfy Cloud MCP server. Cloud-only — connects to the hosted service, not a local ComfyUI install. Use for any "generate/edit an image", "make a video", "text to 3D", or ComfyUI workflow task.
Fetch YouTube video subtitles/captions using Felo YouTube Subtitling API. Use when users ask to get YouTube subtitles, extract captions from a video, fetch transcript by video ID or URL, or when explicit commands like /felo-youtube-subtitling are used. Supports language and optional timestamps.
Write high-quality GPT-Image-2 prompts with strong structure, layout logic, and production-ready visual direction. Use this skill whenever the user asks for image prompts, wants better GPT-Image-2 prompting, wants to upgrade a weak image prompt, asks for prompt templates, wants series-style prompt systems, needs prompts for posters, UI screens, information graphics, concept art, brand visuals, editorial layouts, or worldbuilding images, or wants funny, absurd, meme-like, fake-news-style, and highly shareable visual prompts. If the current Hermes environment exposes an image generation tool/model, this skill should also offer to generate an image from the finalized prompt after presenting the prompt.
Single-image generation skill for posters, key art, and editorial illustrations. Defaults to gpt-image-2 but is provider-agnostic — the same workflow drives Flux, Imagen, or Midjourney via the active upstream tooling. Output is one or more PNG/JPEG files saved to the project folder.
Develop a new reusable image house style through guided reference curation, visual hypothesis testing, canonical anchor generation, range validation, and packaging as an installable skill. Use when the user asks to create, discover, formalize, or package an image aesthetic or repeatable visual system. Do not use merely to generate an image in an existing style.
Download videos and audio from 1500+ websites including YouTube, Bilibili, TikTok, Twitter/X, Instagram, Vimeo, and more using yt-dlp. Use when the user wants to download videos, save media from social platforms, download with specific resolution (720p/1080p/4K), get subtitles, or download entire playlists. Triggers on requests like 'download this video', 'save this YouTube video', 'download in 1080p', 'download with subtitles', 'download this playlist'.
A single-frame motion-design composition with looping CSS animations — rotating type ring, animated globe, ticking timer, parallax labels. Renders as a hero video poster you can hand straight to HyperFrames or any keyframe-based exporter. Use when the brief asks for "motion design", "animated hero", "loop", "video poster", "title card", or pairs Open Claude Design with HyperFrames for a kinetic export.
Revise Chinese writing to remove AI-flavored, template-like expressions while preserving meaning, facts, and author voice. Use when the user asks to 去除AI味, 降低AI感, 改得更像人写, 润色小红书/公众号/随笔/文案/评论/邮件/小说正文, or specifically remove patterns like “不是...而是”, “先...再”, “真正...的是”, “这次只看”, “X很简单：”, mechanical contrast/sequence framing, over-neat parallel rhythm, repeated route markers, and unexplained rhetorical questions at the end.
Agent Skill for writing production-ready SVG markup — icons, logos, favicons, illustrations, flowcharts, charts, progress rings, sprites, CSS masks, and animated UI graphics. Use when the user asks for SVG, vector graphics, inline graphics, icon sets, data viz, or scalable web assets, or mentions this skill. Prefer over image generation when scaling, theming, CSS control, or code integration matters.
Turn a raw talking-head recording into a finished 1080x1920 Short, Reel, or TikTok with B-roll, motion graphics, captions, and music. Show the plan in chat, then iterate on drafts based on the user’s frame-by-frame comments. Use when someone wants to edit a raw recording into a vertical video or add captions, B-roll, or motion graphics to one
Create visual analogies by mapping relational structure from familiar domains onto unfamiliar concepts using spatial relationships to make abstract patterns concrete. Covers static diagrams AND animated video storytelling (camera choreography, race comparisons, pacing). Use when explaining complex concepts, creating analogies, designing diagrams, creating explainer animations, or revealing system structure.
Expert in real-time game animation systems including skeletal animation, blend trees, state machines, inverse kinematics, root motion, procedural animation, and animation retargeting. Specializes in creating fluid, responsive character animation that balances visual quality with performance constraints. Use when "animation system, state machine, blend tree, skeletal animation, inverse kinematics, IK system, root motion, animation blending, character animation, animator controller, motion matching, animation retargeting, foot IK, look at IK, aim offset, animation montage, animation notify, additive animation, animation layers, procedural animation, animation, game-dev, character, state-machine, unity, unreal, godot, skeletal, rigging, motion" mentioned.
Audio generation skill—jingles, beds, voiceover, and sound effects. Routes music requests to Suno V5 / Udio / Lyria, speech to MiniMax TTS / FishAudio / ElevenLabs V3, and SFX to ElevenLabs SFX or AudioCraft. Output is one MP3/WAV file saved to the project folder.
Create and check long-running video material evaluation tasks. Use this skill when the user wants to submit videos for evaluation, check an existing video evaluation task list, or fetch the result of a previously created video evaluation task. This skill is not a general-purpose video upload skill because upload is allowed only as an internal step of task creation. Authentication uses an API key passed as an Authorization bearer token.
Communicate ideas and information with purpose, clarity, and integrity, in any form or tradition and at any length, to an audience of one or many, addressed or not. Works out with the user who the piece is for, what brought it on, and where it should take its audience, then drafts and checks the piece against that, keeping the user's meaning theirs. Use when the user asks for help with writing, commentary, or communication; when they point out AI slop; when they ask things like "help me say", "write for [a specific audience or context]", "polish [comments, sentences, artifacts]", "write something for the poster", "help me write a novel, a handbook, or a long report", "keep this long piece consistent across chapters", or the like.
Generate article cover images with 5 dimensions (type, palette, rendering, text, mood). Supports cinematic (2.35:1), widescreen (16:9), and square (1:1) aspects. Use when user asks to 'generate cover image', 'create article cover', or 'make cover'.
Use when creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Triggers include requests for generative art, algorithmic art, flow fields, particle systems, or any art created through code. Always create original algorithmic art — never copy existing artists' work to avoid copyright violations.
This skill should be used when the user wants to view, review, rate, organize, search, or export images / AI-art generations with the Cull app. Trigger on "show me these images", "review this batch", "open these in Cull", "rate / shortlist / collect these", "find similar images", "make a smart collection", "run a quality pass", "export the keepers", "publish this collection". Works via the `cull` CLI by default (no MCP required); the `mcp__cull__*` tools are optional for richer interactive control.
Use when creating beautiful visual art in .png and .pdf documents using design philosophy. Triggers include requests to create posters, pieces of art, designs, or other static visual pieces. Always create original visual designs — never copy existing artists' work to avoid copyright violations.
Train custom AI models (LoRA) on fal.ai — personalize image generation for specific people, styles, objects, or video generation. Use when the user requests "Train model", "Train LoRA", "Fine-tune", "Custom model", "Train on my images", "Portrait training".
Complete fal.ai video-to-video system. PROACTIVELY activate for: (1) Kling O1 video editing, (2) Sora Remix transformation, (3) Video upscaling, (4) Frame interpolation, (5) Style transfer (anime, painting), (6) Object replacement/removal, (7) Color correction, (8) Video enhancement pipelines. Provides: Edit types (general/style/object), upscaling options, style keywords, enhancement workflows. Ensures consistent video transformation without flickering.
Complete FFmpeg command syntax reference covering option ordering, input vs output options, stream specifiers, and position-sensitive options. PROACTIVELY activate for: (1) Command syntax questions, (2) Option placement issues, (3) Input vs output option confusion, (4) Stream specifier syntax, (5) -ss/-t/-to position questions, (6) Global vs per-file options, (7) Multiple input/output handling, (8) Option order errors. Provides: Correct option placement rules, input-only vs output-only options, position-sensitive option behavior, stream specifier syntax, common mistakes and fixes.
Complete live streaming and protocol system for FFmpeg 7.1 LTS and 8.0.1 (latest stable, released 2025-11-20). PROACTIVELY activate for: (1) RTMP streaming to Twitch/YouTube/Facebook, (2) HLS output and adaptive bitrate (ABR), (3) DASH streaming setup, (4) Low-latency streaming (LL-HLS, LL-DASH), (5) SRT protocol configuration, (6) WebRTC/WHIP sub-second latency (FFmpeg 8.0+), (7) Protocol conversion (RTMP to HLS), (8) Multi-destination streaming, (9) nginx-rtmp integration, (10) Docker streaming services. Provides: Platform-specific stream commands, ABR ladder examples, encryption setup, latency optimization, WHIP authentication, production patterns. Ensures: Reliable live streaming with optimal quality and latency.
AI film director for scene blocking, shot composition, visual storytelling. Use when: directing, blocking, shot design, coverage, cinematography. Generates image prompts for scenes. Keywords: film, directing, blocking, shots, visuals.
Hilarious LinkedIn profile roasting skill that delivers savage but friendly comedy burns based on someone's profile, activities, skills, experience, or lack thereof. Use when users ask to roast a LinkedIn profile, make fun of someone's professional presence, or want a humorous critique of LinkedIn content. Triggers on requests like "roast this LinkedIn", "make fun of this profile", "roast me", "give me a LinkedIn roast", "flame this profile", or any request for comedic mockery of a professional profile.
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.