Skills
100,928 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Draft, audit, or revise point-by-point reviewer response letters for Nature-family manuscript revisions. Use when the user provides reviewer comments, editor decision letters, revision notes, response drafts, or asks how to respond to major/minor revision requests, rebuttal letters, response to reviewers, peer-review reports, 审稿意见回复, 逐点回复, 修回信, 大修回复, 小修回复, or 如何回复 reviewer
Rotate old AI run folders and standalone plans into one verified .tgz batch archive beside the remaining runs. Keeps the last 14 days unpacked by default, with configurable retention, exclusions, a preview, and rotation delivered through om-auto-create-pr. Use to archive old .ai/runs context; active or uncertain runs stay unpacked.
Route design-generation work into the right Open Design workflow mode — prototype, deck, document, or media artifact. Use when the user needs local-first UI prototype generation, presentation deck creation, or design-system-aware HTML/PDF/PPTX artifacts via locally-installed coding agents. Supports 72 built-in design systems, 5 visual directions, 93 media prompt templates, and multi-format export. Triggers on: open-design, local design tool, prototype generation, design deck, design artifact, design agent workflow, open design prototype, design system artifact, UI prototype generation.
Convert entire PDF documents to clean, structured Markdown for full context loading. Use this skill when the user wants to extract ALL text from a PDF into context (not grep/search), when discussing or analyzing PDF content in full, when the user mentions “load the whole PDF”, “bring the PDF into context”, “read the entire PDF”, or when partial extraction/grepping would miss important context. This is the preferred method for PDF text extraction over page-by-page or grep approaches.
Use when driving Xiaohongshu followers to private domains like WeChat, building offline communities, implementing follower retention strategies beyond platform, or creating sustainable audience ownership independent of platform algorithms
Protracted strategy: For long-term, complex tasks that cannot be won quickly, objectively assess the situation, divide the effort into defense / stalemate / counter-offensive stages, determine the current stage, set transition conditions, and identify local offensive opportunities even during difficult stages. Trigger when a task spans multiple sessions or iterations, resources are temporarily insufficient, or either “wanting to give up” or “going all in” arises; do not trigger when the task can be completed in the current session.
Senior QMS audit expert for internal and external quality management system audits. Provides ISO 13485 audit expertise, audit program management, nonconformity identification, and corrective action verification. Use for internal audit planning, external audit preparation, audit execution, and audit follow-up activities
Rebuild an existing site's pages with byte-faithful content on a separately-defined donor design system — another live site, or local static HTML prototypes (Figma donors are contract-defined, not yet implemented). Content fidelity is gated byte-level (text, ordered images, SEO metadata); design application is flexible (content re-laid-out onto donor modules). Use when the user says “reskin my site with this design”, “apply this design system to my content”, “restyle my site to look like <other site>”, “new design, same content”, “rebrand my site using these prototypes”, or names a content site plus a design donor. NOT for redesigning from intent (that's the stardust core extract/direct/prototype chain) and NOT for keeping the current design while migrating (that's the replica flow)
Rive is the interactive engine for games, apps, and UI. Use it as your whole engine, or start inside what you already use: native apps, Unity, Unreal or your own C++ engine. As a full engine it ships to web, mobile, desktop, TVs, cars, PS5, Xbox and Switch, no other engine required. The core runtime is tiny and MIT open source, with no royalties, rev share or runtime fees. One Rive file holds UI, vector, raster and 3D graphics, with motion, interaction and logic. It can load any data, asset or file at runtime, including other Rive files. Coding agents build Rive files from text with the Rive CLI. People refine them in the Rive Editor. Build whole apps in Rive when they need unique graphics or motion to stand out. The CLI ships them as iOS, Android and web apps. Any other app can use Rive where it should stand out: onboarding, characters and charts. Use it for HUDs and menus, instrument clusters, TV apps, ads and landing-page heroes. In games, vehicles and TVs, build every screen in Rive, static ones too.
Systematic investigation of metric changes and anomalies. Use when a metric changes unexpectedly, investigating business metric drops, explaining performance variations, or drilling into the drivers of aggregated metrics.
Find keywords competitors rank for that the target property does not, and prioritize them by opportunity. Uses Ahrefs MCP for keyword and competitor data. Use this skill when planning content investment, identifying quick wins, building a content calendar against a competitor set, or scoping a market entry. Triggers on keyword gap, content gap, competitor keywords, opportunity keywords, what should we target, where are competitors winning, keyword opportunity. Also triggers when planning content for a new market or after losing organic share to a specific competitor
Creates, updates, lints, and publishes portable agent skills for Claude Code, Codex, and OpenCode, including repo-local .agents/skills sources, Claude compatibility symlinks, cross-agent SessionStart refresh hooks, and npx distribution. Use when (1) writing or editing a skill, (2) linting with agnix, (3) publishing changes, (4) renaming or deleting a skill, or (5) deciding skill layout, scope, frontmatter, discovery, or refresh automation. Triggers include "new skill", "update skill", "repo skill", "skill not loading", "npx skills", "skills update", "Codex skill hook", "lint skill", "agnix", "SKILL.md".
“Make it sound human”: a Traditional Chinese editing skill that removes AI-like phrasing. Reviews and revises text to remove AI-like wording, correct China-specific terms and half-width punctuation, and make it read as if written by a person. Triggers when the user says “remove the AI flavor,” “make it sound human,” “this sounds so AI,” “make it more natural,” “polish this and remove the AI feel,” or asks to check the tone of outward-facing text such as newsletters, social posts, sales pages, course copy, customer support replies, presentations, announcements, and emails before sending. Do not trigger for literal translation, imitating a specific brand voice, fact-checking (when not a style issue), code/log/config files, or requests to “polish it in Raymond’s voice” (that is for the content-writing skill; this skill only removes AI-like phrasing and does not add a personal style).
A collection of Agent Skills for the Stitch MCP server: generate high-fidelity UI screens, create multi-page websites from a single prompt, produce DESIGN.md documentation, enhance vague UI prompts, convert designs to React/shadcn-ui components, and generate walkthrough videos via Remotion. Use when the user needs AI-assisted UI design generation, prompt refinement, or screen-to-code workflows. Triggers on: stitch, stitch-design, stitch-loop, enhance-prompt, react-components, remotion, shadcn-ui, screen generation, ui generation.
Video downloading, stitching, title slides, and YouTube descriptions. Use this skill when downloading videos from Google Drive, creating title slides, stitching multiple videos together, or generating YouTube descriptions with timestamps. Triggers on video course creation, video editing, video compilation, or YouTube upload preparation.
Generate images and text via Doubao Web using browser cookies and Playwright UI automation. Use when a user asks for Doubao Web image generation, prompt-to-image automation, text replies, preset aspect ratios (e.g. 小红书封面), or session-based multi-turn workflows.
Use when you need to design, review, or improve validation in Micronaut applications — including Bean Validation on @Controller methods, @Body @Valid, query/path parameter validation, @ConfigurationProperties validation, custom constraints, nested DTO validation, and ExceptionHandler mapping for constraint violations. This should trigger for requests such as Add validation support in Micronaut; Review Micronaut validation rules; Improve request validation in Micronaut REST APIs; Add custom validation constraints in Micronaut; Validate Micronaut configuration properties. Part of Plinth Toolkit
Discover and reconcile Alibaba Cloud product catalogs from Ticket System, Support & Service, and BSS OpenAPI; fetch OpenAPI product/version/API metadata; and summarize API coverage to plan new skills. Use when you need a complete product list, product-to-API mapping, or coverage/gap reports for skill generation.
Applies Agency brand colors and typography to artifacts including presentations, SVG graphics, documents, and web interfaces. This skill should be used when brand colors, visual formatting, neobrutalism style, or Agency design standards apply. Keywords - branding, corporate identity, visual identity, styling, brand colors, typography, visual formatting, visual design, neobrutalism.
Create actionable buyer personas using Adele Revella's 5 Rings of Buying Insight methodology—based on real buyer stories, not demographics. Use when: understand why buyers buy (or don't buy) your solution; create personas that drive action rather than gather dust; align marketing and sales around real buyer insights; develop messaging that resonates with actual decision-makers; map content to the buyer's journey
Designing interactive calculators (ROI calculators, pricing estimators, savings projections, mortgage calculators, custom assessments) that deliver real decision-support value while serving as lead magnets and qualified-traffic generators. Honest about vanity-calculator (no real value), lead-trap (hides the answer behind email), and transparent-decision-tool (gives the result and earns the email through tiered value) patterns. Triggers on calculator, ROI calculator, pricing estimator, savings calculator, custom calculator, interactive tool, decision tool, financial calculator. Also triggers when an audience needs a calculation-driven lead magnet, when a vanity calculator is producing leads but no qualified ones, or when a calculator is being scoped for the first time.
Superseded by the web CLI: attach to an already-running Chrome with `web browser attach <cdp-url>`, then drive it with `web browser eval|run|cdp|screenshot`. Use when the user asks to attach to, connect to, or control an existing Chrome running with --remote-debugging-port
Shoot software UI like a film. A sequence of hero shots, each a steep, oblique, extreme close-up of the interface floating on a stage (a black void for dark products, a high-key studio for light ones), with a razor-thin focal plane, a slow motivated camera drift, and UI that acts on its own, hard-cut on the action, ending on one line of copy and a still logo. It also shoots stills, the same look frozen into single photographs. Use when asked for a launch video, product film, feature trailer, teaser, promo clip, hero shot, or any "cinematic" footage of an app, UI, or website, and also for a hero image, key visual, product photo, poster frame, launch-post image, Open Graph or social card, or any "cinematic" still of a UI. This skill covers camera, framing, light, and edit. It is stack-agnostic.
Expert CIS Controls v8 (CIS Top 18) advisor — implementation group scoping (IG1/IG2/IG3), control gap assessments, safeguard-level guidance, asset inventory, software inventory, data protection, secure configuration, account management, access control, continuous vulnerability management, audit log management, email and web browser protections, malware defenses, network infrastructure management, network monitoring and defense, application software security, incident response, penetration testing, and CIS Controls mapping to NIST CSF, ISO 27001, SOC 2, and CMMC. Use for any question about CIS Controls, CIS Benchmarks, Implementation Groups, or prioritized cyber hygiene for any organization size.
Generates a comprehensive client health overview across all accounts. Reads CRM data, support tickets, usage metrics, billing, and engagement logs. Calculates health scores, trend direction, and RAG status per client. Outputs a sorted risk report with recommended actions.
Call use_skill as your FIRST and ONLY action when the user asks to DELETE, REMOVE, or GET RID OF an existing run configuration for Mule applications. Use this ONLY for deleting/removing configurations, NOT for creating or editing. Trigger phrases include "delete config", "remove config", "get rid of config", "delete all configs", "remove all my run configs", "clean up configs". When you call use_skill, it must be the only tool call in that response.
Generate AI images from text prompts via the free z-image-turbo model (Gitee AI serverless API, ~100 free 2K images/day with a free token). One-time setup stores the token locally; then generate PNG images in any size (multiple of 8), with negative prompts and step control. Companion web app with 18000+ prompt library and 12 preset styles: https://freeimg.shenzjd.com. Use for generating, drawing, or creating images (icons, covers, illustrations, posters). Keywords: image generation, text-to-image, AI image, z-image, z-image-turbo, Gitee AI, free image, 画图, 生成图片, 文生图, AI 绘画, 免费图片生成, freeimg-z-image.
Use when integrating, generating, refactoring, reviewing, or debugging Make App record-list sorting with @qfei-design/make-app-sort. Triggered by sorting, advanced sorting, multi-field sorting, sort priority, ascending/descending, drag-and-drop sort conditions, table/header/column sorting, openWithField, sortable capabilities, Entity Preset sort save/load/echo, records sort payloads, or sorting tests. Covers one integrated toolbar, CanvasTable header, Entity Preset, and Service sorting flow. When make-app-actions is present, a successfully applied sort must clear its selection and invalidate pending action work; draft edits and failures preserve selection. Does not own page shell/layout, CanvasTable rendering internals, Service route implementation, permission policy, auth, runtime packaging, DSL modeling, Make CLI execution, npm package internals, or grouping
Use when preparing or updating the local Make development environment before development. Triggered by Make 环境安装, Make 环境初始化, 更新 Make 环境. Distinguishes new Apps from existing Apps whose runtime must be preserved; does not manage Make resources, deploy Apps, or write PRD, DSL, Service, or UI code; use makecli for resource/deploy operations and the owning skills for implementation.
Look up details of an existing hotel booking. Use this skill when the user wants to check the status of their reservation, see check-in instructions, verify booking details, or asks "what's the status of my booking", "show me my reservation", "when is my hotel check-in", "did my booking go through". Requires a booking ID, the last name, and the email on the booking.
Convert a legacy UI5 freestyle JavaScript app (sync bootstrap, jQuery.sap.*, ES5, sap_belize) into a modern UI5 1.147 TypeScript app with sap.f.FlexibleColumnLayout, typed event handlers, ES modules, BaseController, and sap_horizon — with 5 documented critical traps up front. Use when asked to "modernize this UI5 app", "convert to UI5 TypeScript", "upgrade jQuery.sap to modern UI5", or "migrate freestyle UI5 to 1.147".
Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote execution. Use when needing to run interpretability experiments on massive models (70B+) without local GPU resources, or when working with any PyTorch architecture.
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.