Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Creative & media
Blender 5.x animation and rigging — keyframes, FCurves, layered actions, drivers, constraints, armatures, IK/FK, shape keys, NLA editor, and bone collections via Python (bpy). Includes 5.1 changes (Smooth Gaussian FCurve modifier, Apply to Basis, layered action performance).
Create AI-optimized style guides for brand or personal voice. Extracts patterns from writing samples or AI-draft/final-sel calibration pairs, producing a reusable guide with tone rules, structure patterns, signature moves, anti-filler checklists, anti-pattern blacklists, and examples. Output is optimized for LLM consumption (Claude Projects, system prompts). Not for editing content (use copy-editor).
Enhanced speech recognition (ASR Pro): transcribe audio/video to text with enhanced analysis such as noise reduction and speaker diarization. Triggered when the user wants to transcribe recordings/videos, generate subtitles, distinguish speakers, or perform enhanced speech analysis.
Removes unwanted visual elements from videos using AI-powered inpainting via Volcengine LAS. Video watermark removal, subtitle removal, logo removal, and text overlay removal — erases and cleans up watermarks, hardcoded subtitles, logos, text, and other fixed-region artifacts and obstructions from video footage. Video inpainting, video repair, video restoration, and video cleanup. Supports specifying exact bounding boxes for targeted removal and erasure. Handles videos up to 4 hours and 30GB. Use this skill when the user wants to remove watermarks from videos, erase hardcoded subtitles/captions, delete logo or text overlays, clean up video footage by removing unwanted objects, repair or restore videos by removing obstructions, do video inpainting on fixed regions, or remove visual artifacts from video content.
Query the Chinese classical poetry API service (chinese-poetry-api, repo: https://github.com/palemoky/chinese-poetry-api) — 370k+ poems from Tang/Song/Yuan dynasties, with authors, dynasties, poetry types, full-text search, and random poem drawing. Use whenever the user asks to look up, search, or list classical Chinese poetry (古诗/唐诗/宋词/元曲), a specific poet's works, random poems, 飞花令 (single-character draws), or poetry metadata — even casual queries like "帮我找一首李白的诗" or "来一首五言绝句". Covers every REST endpoint: health, stats, poems (filtered lists), search, random, authors, dynasties, types. All output is raw server JSON, passed through unchanged.
Create a new tech blog post on 聆.tw (琳聽智者漫談), your AI-driven tech blog. Use when the user wants to write a new blog post, create an article, draft a tech article, or publish content on 聆.tw. Triggers on requests like "write a blog post", "create an article about X", "draft a post on 聆.tw", or "help me write about X for the blog". This skill handles the full workflow: repo setup, content creation, following strict editorial guidelines, build check, and submitting a pull request.
Douyin account video batch downloader — enter a Douyin ID and it automatically pulls all works of the account and resolves watermark-free download links, supporting one-click batch download to local storage. Use cases: breaking down and frame-by-frame studying competitor account content, saving all works of a favorite creator to avoid deletion, batch transferring material for secondary editing, saving your own account's works as local backup against loss, stockpiling course/tutorial videos for offline repeated viewing, providing watermark-free source files when a client requires them, and quickly pulling a partner's works for due-diligence analysis before collaboration. Trigger words: Douyin homepage download, Douyin video download, Douyin batch download, Douyin watermark-free download, download Douyin works, Douyin account video extraction, Douyin video saving, Douyin watermark removal, Douyin video backup, competitor video download.
Generate ElevenLabs text-to-speech audio from scripts or inline text using local voice profiles. Use when the user asks for ElevenLabs, text-to-speech, TTS, narration, voiceover, speech audio, or voice generation; load voice names, voice ids, emails, owners, and account-specific defaults only from local config outside the skill.
Research-backed principles for writing prose that avoids AI tells. Apply when writing articles, blog posts, emails, marketing copy, social media, or any prose content. Covers vocabulary, structure, tone, rhythm, and craft techniques that make writing feel authentically human. Not for code, commit messages, or technical documentation.
Remove traces of AI generation from text so it sounds more natural and human-written. Use for editing or reviewing text. Trigger when the user mentions "removing AI flavor", "humanizing", "more like plain speech", "sounds unlike AI", or needs to rewrite AI-generated text to sound more natural. Based on the Wikipedia "Signs of AI writing" guide v2.5.1, it detects and fixes 29 AI patterns.
Break down an article and extract reusable material (opinions, data, cases, quotable lines, analogies, methodologies), label it by category, and store it in the material library. Use when the user says "拆解素材", "入库", "收集这篇文章", "提取素材", "分析这篇文章的要点", or feeds in an article hoping to retain valuable content. Even if the user merely shares an article and says "这篇不错", consider whether it should be stored.
Japanese style guide and AI-tone rubric for mizchi-authored blog posts (zenn / dev.to), calibrated on 18 articles / 937 sentences / 766 paragraphs. Scores a draft structure-first — paragraph granularity, heading form, code cadence, closing move — because marker rates (口語・文末・自分) can be hit exactly while the text still reads as pastiche. Use after drafting an article during the style-matching revision loop.
Applies motion design principles to create emotionally-driven, technically sound animations and transitions. Provides timing, easing, choreography, and Disney animation principles adapted for UI. Use when creating animations, transitions, micro-interactions, loading states, page transitions, scroll-triggered effects, or any motion work. Works with CSS, Framer Motion, GSAP, Lottie, Spring, or any animation system.
Build and validate Open Design handoffs from the real Darkbone Archer runtime — game UI (upgrade-only preservation contracts) AND VFX/skill/RIG design packages. Use whenever Codex must prepare, regenerate, or audit an Open Design handoff; feed an existing Claude Design handoff zip into Open Design; collect current screenshots, animation videos, keyframes, source anchors, assets, responsive states, or preservation boundaries; or prevent a standalone concept from deleting existing identity, motion, interaction, copy, or data that it does not show.
Dispatcher for the creative suite: diagnoses a multi-asset creative request with a few questions, then routes it through the right sequence of creative skills, naming the artifact each step passes to the next, and wires in the FAL.ai automation (single model fal-ai/nano-banana-pro via docs/creative_cli.py) when available. Use when the user asks for several creative deliverables at once or an outcome that implies them — 'assets for my product launch', 'set up my brand visuals', 'I need a logo, product shots, and social posts', 'a week of content', 'rebrand everything' — or when it's unclear which creative skill applies. Produces a Creative Production Plan: chosen route, ordered skill sequence with hand-off artifacts, generation checklist, and asset output map. For a single obvious asset, skip this and use the specific skill directly.
Monsters, dragons, world reveals. Each entity gets its own definition block, the shots are cut by timecode, and scale is bought with reference objects and low angles rather than words like massive. Use when the user wants a Seedance 2.5 / 2.0 video prompt for this kind of clip (Epic fantasy and sci-fi spectacle; 中文触发词: 奇幻科幻大场面, 奇幻, 科幻, 巨龙, 怪兽, 史诗), or asks how published cases of this type were prompted. Read references/cases.md before drafting.
One person, one look, a handful of places. A head-to-toe appearance lock carries the whole clip, and each scene gets one location, one gesture and one kind of light. Use when the user wants a Seedance 2.5 / 2.0 video prompt for this kind of clip (Fashion lookbook and portrait film; 中文触发词: 时尚 lookbook 与人像写真片, 时尚, 穿搭, 人像, 写真), or asks how published cases of this type were prompted. Read references/cases.md before drafting.
The animal is the lead and a phone is the only camera. Lock the count to exactly one, keep the animal behaving like an animal, and let the payoff come from it closing in on the lens. Use when the user wants a Seedance 2.5 / 2.0 video prompt for this kind of clip (pets and animals as the lead; 中文触发词: 宠物动物当主角, 宠物, 猫, 狗, 动物), or asks how published cases of this type were prompted. Read references/cases.md before drafting.
Recommend AI short drama, micro drama and manju (漫剧) skills from a hand-checked list of 90+ projects, matched to the user's stage (novel adaptation, episode outlines and hooks, script writing, review, storyboards, character consistency, video prompts, production), video model (Seedance/即梦, Kling/可灵, MiniMax, Wan, Veo, Sora, ComfyUI) and agent (Claude Code, Codex, OpenClaw/龙虾, Qwen, Doubao, WorkBuddy). Use when someone asks which skill or tool to use for an AI short drama or 短剧/漫剧, or how to install one. Helps you choose from a manually verified list of short-drama and manju skills: novel adaptation, episode outlines and hooks, scripts, review, storyboards, character consistency, video prompts, and video production.
Steve Jobs' thinking framework and way of expressing himself. Based on deep research into 30+ primary sources, including the authorized Isaacson biography, the Stanford speech, the Lost Interview, the D Conference series, and Make Something Wonderful, it distills 6 core mental models, 8 decision heuristics and a complete expression DNA. Purpose: to serve as a conversational thinking advisor — analyzing your products, decisions and ideas from Jobs' perspective, creating intellectual friction and sparking inspiration in back-and-forth dialogue, and helping you internalize "Jobs-level" thinking as your own ability. Use when the user mentions "from Jobs' perspective", "what would Jobs think", "Jobs mode", "steve jobs perspective", "switch to Jobs", "if Jobs", "Jobs thinking", or "expression DNA". It should also trigger even if the user only says "help me think about it from Jobs' angle", "what would Jobs do", or "switch to Jobs".
Analyze an article's writing-style features and extract style dimensions into a style material library. Can blend multiple style materials to generate or update the main style profile (my_style.json). Use when the user says "分析风格", "提取写作风格", "学习这个语气", "分析我的文风", "吸收这个风格", or "更新我的风格". Even if the user merely shares an article and expresses interest in its style, consider using it.
Choose, configure, and optimize Three.js mesh materials: basic, Lambert, Phong, Standard, Physical, toon, points, lines, and shader-backed surfaces; PBR maps, transparency, environment reflections, cloning, and disposal. Use when styling meshes, tuning PBR, fixing transparency or material sharing bugs, or reducing material cost. Triggers on: Three.js material, MeshStandardMaterial, MeshPhysicalMaterial, PBR, roughness, metalness, transparency, environment map, mesh surface, material clone.
Use when a 3D asset must reach a web scene within budget: source models and textures, manage the glTF pipeline (KTX2, meshopt, Draco), loaders, colour spaces, and licences. NOT for scene code (web3d-runtime), motion (web3d-animation), an in-flight Foundry order (asset-foundry), or Quest store assets (xr-dev). Opt out: no web3d, без web3d. Triggers: "compress this GLB" / "сожми модель", "KTX2", "meshopt", "the model is too heavy" / "модель слишком тяжёлая".
A WeChat official account article generator designed for Dify workflow case sharing. It follows the Xiao Hui Hui official account writing guidelines, automatically generates complete Dify case articles with an introduction, workflow creation and summary, includes detailed node configurations, plugin installation steps and code examples, and supports automatically generating images and uploading them to the Tencent Cloud COS image hosting service.
Use this skill when a user names or describes a visual aesthetic (Y2K, vaporwave, brutalist, cottagecore, dark academia, etc.), wants to compare or disambiguate aesthetics, or needs formal characterization before producing design tokens or assets. Provides the 7-dimension framework (palette, type, texture, shape, motion, spatial conventions, cultural markers) and a curated dictionary of 168 major aesthetics. Foundation for aesthetic-application. Do not use for component implementation, layout, or accessibility decisions
AI Talking Head — Generate Talking Head Videos from Text with AI Presenter. Turn any script into a talking head video without hiring a spokesperson or setting up a camera. Paste your text, pick a presenter style, and the AI generates a video with natural lip sync, gestures, and pacing. Use it for product walkthroughs, course intros, investor updates, FAQ pages, or internal training. Edit the result in chat: swap the background, adjust delivery speed, add subtitles, or insert the talking head clip into a longer video alongside screen recordings and slides. No green screen, no teleprompter, no retakes. Supports mp4, mov, webm, mkv output.
Quantitative detection and scoring of AI-ness in Chinese web fiction. Use when users want to "give the text an AI-ness score / quantify how AI-like it is", need to know "which sentence exactly the problem is in", want to check metaphor density, consecutive identical sentence patterns, sentence-length variation, paragraph distribution, repeated high-frequency emotion words, or piled-up extreme words, or want to locate the AI-ness hotspots before revising. Produces an AI-ness index (0–10) + measured results for eight metrics + a deduction location checklist.
Use when managing Alibaba Cloud ApsaraVideo Live resources and workflows via OpenAPI/SDK, including live domain configuration, stream ingest and playback setup, recording/transcoding templates, monitoring queries, and live stream operations.
Use when transforming video style with DashScope video-style-transform model. Use when converting videos to artistic styles such as Japanese manga, American comics, 3D cartoon, Chinese ink painting, paper art, or simple illustration via the video-synthesis async API.
Use when editing videos with DashScope Wan 2.7 video editing model (wan2.7-videoedit). Use when implementing video style transfer, instruction-based video editing with optional reference images, or video content modification via the video-synthesis async API.
Write articles, guides, blog posts, tutorials, newsletters and other long-form content. This content has a distinctive tone extracted from the provided examples or brand guidelines. Use this skill when the user needs polished copy longer than a paragraph and requires consistent tone, structure and credibility.
Based on existing content assets such as articles, columns, course lectures, verbatim transcripts, interviews, courseware, meeting minutes, case materials, PDF text, Word documents, and notes, determine whether they are best turned into a book, booklet, course, article series, practical handbook, or knowledge base, and output concise planning advice. This skill should be used when the user mentions scenarios such as "organize existing content into a book", "judge whether the material suits a book or a course", or "recombine articles/live-stream transcripts/course transcripts/notes into a knowledge product".
Use this skill to create single-voice audio content like audiobooks, voiceovers, narrations, jingles, and audio ads. Triggers: "create audiobook", "generate voiceover", "narration", "audio ad", "radio ad", "jingle", "brand audio", "sonic logo", "text to audio", "read this aloud", "audio guide", "meditation audio", "soundscape" Orchestrates: narration/TTS, background music, and audio assembly. NOTE: For conversations/dialogues, use podcast-producer instead.
Audit and rewrite content to remove AI writing patterns ("AI-isms"). Use this skill when asked to "remove AI-isms," "clean up AI writing," "edit writing for AI patterns," "audit writing for AI tells," or "make this sound less like AI." Supports a detection-only mode that flags patterns without rewriting.
How to write H3 prompts for brand/product promo shorts: brand facts, narrative spine, precise beats, outputting prompts that can be fed to MiniMax H3 in segments. Applies to launch/website/social media promo prompts. Not applicable to unauthorized imitation of logos or long scripts with fabricated selling points.
Author a PostHog canvas with semantic HTML, CSS, and direct browser APIs — documents, articles, generative graphics, 2D canvas and WebGL experiences, and focused experiments where React components add no useful structure. Use after building-canvases has routed a canvas request to a plain-HTML/browser-API implementation. Covers the thin component wrapper the current runtime requires, styling and theming without Quill, drawing surfaces, and animation/cleanup patterns.
Drive a local ComfyUI install for image, video, and audio generation via its HTTP API. Use WHENEVER generating, rendering, or editing images/video/audio/hero assets with ComfyUI, Z-Image, Ideogram, FLUX, LTX, Wan, or when building, parameterizing, or running ComfyUI workflows. Covers the API client, workflow JSON format, model patterns, dual/multi-GPU placement, the MCP driver, in-graph Claude nodes, and VRAM coordination.
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.