Skills
101,323 packaged capabilities and playbooks for agents. Install any of them with one command.
All
For all content creators who want to turn "gut feeling" into calibratable predictions. The methodology is general — the loop of scoring → blind prediction → T+3d retrospective → evolving the rubric applies to any content that can be quantified (views / reads / listens / clicks). The rubric is the content of the loop, not the loop itself — it currently ships with one opinion-video rubric (fitted against 25+ videos from a reference creator); other formats can start from it and bump the weights. Strongly recommended to import benchmark accounts as the initial signal source (/cheat-learn-from). Trigger words: "initialize" / "score this one" / "start prediction" / "published" / "retrospective" / "upgrade rubric" / "recommend topics" / "grab trends" / "status" / "find benchmarks" / "learn from". First-time use must run /cheat-init first.
Internal downstream skill for ctf-sandbox-orchestrator. CTF-sandbox workflow for CTF web, API, SSR, frontend, queue-backed app, and routing challenges. Use when the user asks to inspect a site or API, follow real browser requests, debug auth or session flow, trace uploads or workers, find hidden routes, or explain why frontend and backend behavior diverge under sandbox-internal routing. Use only after `$ctf-sandbox-orchestrator` has already established sandbox assumptions and routed here.
When the user wants to create competitor comparison or alternative pages for SEO and sales enablement. Also use when the user mentions “alternative page,” “vs page,” “competitor comparison,” “comparison page,” “[Product] vs [Product],” “[Product] alternative,” “competitive landing pages,” “how do we compare to X,” “battle card,” or “competitor teardown.” Use this for any content that positions your product against competitors. Covers four formats: singular alternative, plural alternatives, you vs competitor, and competitor vs competitor. For sales-specific competitor docs, see sales-enablement.
Build modern, composable, and accessible React UI components following the components.build specification. Use when creating, reviewing, or refactoring component libraries, design systems, or any reusable UI components. Triggers on tasks involving component APIs, composition patterns, accessibility, styling systems, or TypeScript props.
Docker, Kubernetes, and AWS ECS/Fargate patterns. Triggers on: Dockerfile, docker-compose, kubernetes, k8s, helm, pod, deployment, service, ingress, container, image, ecs, fargate, task definition, ecs service, awsvpc, FARGATE_SPOT, ALB, ecs vs kubernetes.
Use when the user asks to "review this influencer content" or "check if this post meets brand guidelines"; runs a typed C3 ART asset gate, checks disclosure and claim-integrity vetoes, and writes constructive revision feedback. Not for drafting the brief — use brief-generator; not for partnership terms — use contract-helper.
Create or update a Coral source spec YAML for a custom HTTP API or local dataset. Use when authoring a standalone source for `coral source add --file`, or when adapting that spec into a Coral repo source under `sources/core` or `sources/community`.
Generate custom design system rules for the user's codebase. Use when the user says "create design system rules", "generate rules for my project", "set up design rules", "customize design system guidelines", or wants to establish project-specific conventions for Figma-to-code workflows. Requires a connection to the Figma MCP server.
Step-by-step workflow for creating complete Power BI reports from scratch using pbir CLI. Covers model discovery, report creation, page layout, theme setup, visual placement, field binding, filtering, formatting, validation, and publishing. Automatically invoke when the user asks to "create a new report", "build a report from scratch", "make a dashboard", "set up a report with KPIs", "create an executive dashboard", "add pages and visuals to a new report".
Reshapes one Instagram reel into a TikTok and a YouTube Short — adjusting hook, length and captions per platform. Use when the user says "repost this to TikTok", "make this a Short", "cross-post this reel", "reformat for YouTube".
Process JSON with jq and YAML/TOML with yq. Filter, transform, query structured data efficiently. Triggers on: parse JSON, extract from YAML, query config, Docker Compose, K8s manifests, GitHub Actions workflows, package.json, filter data.
Design or restyle DatoCMS plugins so they look and feel native to the DatoCMS UI. Use when users ask to make a plugin match the DatoCMS dashboard, polish plugin config screens, pages, sidebars, panels, modals, forms, tables, empty states, or overall plugin layout structure. This skill owns DatoCMS plugin design-system work, native-look restyling, and UI density or spacing cleanup. Prefer `datocms-react-ui` when a public component exists, and otherwise use raw React and CSS that reproduce DatoCMS spacing, typography, density, color, and interaction patterns without importing private CMS classes.
Working profiles of Saul Bass, Massimo Vignelli, Dieter Rams, Paula Scher, Josef Müller-Brockmann, David Carson, and Paul Rand: key works, principles, and how to apply each in UI work. Use when a design needs a proven approach, a named reference to follow, or a lesson from a specific master.
[devup-ui] Routing-first front door for Devup UI — a zero-runtime CSS-in-JS library whose Rust + WebAssembly preprocessor extracts every style at build time, so no styling JavaScript ships at runtime. Use to adopt/integrate Devup UI, wire the build-time plugin into Next.js/Vite/Rsbuild/Webpack/Bun, write styles with Box/css props or the styled-components-compatible styled() API, set up type-safe theming with devup.json, or migrate off styled-components/Emotion/Tailwind/Panda/vanilla-extract to zero runtime.
Automatically discover software engineering practice skills when working with code review, documentation, pair programming, production debugging, performance profiling, deployment strategies, or software engineering practices. Activates for engineering development tasks.
Expert guidance for the `docker build` command — building container images, tagging strategies, build arguments, image management and inspection. Covers `docker build` complete syntax, `--build-arg` usage, image tagging (`docker tag`), image inspection (`docker inspect`, `docker history`, `docker image ls`), pushing to registries (`docker push`), image pruning (`docker image prune`, `docker system prune`), and Docker BuildKit basics. Distinct from docker-dockerfile (focuses on Dockerfile authoring) and docker-buildx (focuses on multi-platform builds and advanced BuildKit). Use when the user asks about docker build, docker tag, docker push, docker images, docker image management, or how to build and manage Docker images. 使用场景:docker build 构建镜像、docker tag 打标签、docker push 推送镜像、docker images 镜像管理、docker image 命令、构建参数 --build-arg、镜像标签、docker build 命令.
Guidance for integrating Docker into CI/CD pipelines. Covers GitHub Actions (docker/build-push-action, multi-platform builds with buildx, cache-from/to), Jenkins Pipeline (Docker agent, build/push stages), Docker Build Cloud for accelerated builds, layer caching strategies in CI, image tagging strategies for CI/CD, deployment patterns (rolling update, blue-green). Use when the user asks about Docker CI/CD, GitHub Actions Docker, Jenkins Docker pipeline, buildx multi-platform, Docker Build Cloud, or needs to automate Docker builds. 使用场景:docker CI/CD、GitHub Actions docker、Jenkins docker、buildx、多平台构建、Docker Build Cloud、流水线.
Use when writing, reviewing, restoring, trimming, or auditing prose in the deepseek-harness repo, including deciding where documentation or comments are required across Markdown, JSDoc, code and test comments, prompts, descriptions, diagnostics, and CLI or UI strings.
CSS-native animation techniques including Scroll-Driven Animations API, View Transitions, @property rule, visual effects, and micro-interaction patterns. Use when asked about: CSS animations, scroll-driven animations, CSS scroll timeline, view transitions, page transitions, @property, animating CSS variables, clip-path animations, backdrop-filter, mix-blend-mode, CSS masks, CSS-only animations, no-JavaScript animations, micro-interactions, hover effects, accordion animations, or when GSAP is not needed/wanted. Includes decision guide for CSS vs GSAP. IMPORTANT: Always load elite-accessibility alongside this skill for reduced-motion patterns.
When the user wants to create or optimize an email sequence, drip campaign, automated email flow, or lifecycle email program. Also use when the user mentions "email sequence," "drip campaign," "nurture sequence," "onboarding emails," "welcome sequence," "re-engagement emails," "email automation," "lifecycle emails," "trigger-based emails," "email funnel," "email workflow," "what emails should I send," "welcome series," or "email cadence." Use this for any multi-email automated flow. For cold outreach emails, see cold-email. For in-app onboarding, see onboarding.
Use when testing enterprise integrations across SAP, middleware, WMS, or backend systems, validating E2E enterprise flows, testing SAP-specific patterns (RFC, BAPI, IDoc, OData, Fiori), or enforcing cross-system quality gates.
Build automated ETL (Extract-Transform-Load) pipelines for construction data. Process PDFs, Excel, BIM exports. Generate reports, dashboards, and integrate with other systems. Orchestrate with Airflow or n8n.
This skill is triggered when multiple pieces of evidence in a case need to be systematically mapped to legal argument claims. Typical trigger scenarios include: preparing litigation documents (complaint, statement of defense, attorney's argument) and needing to build a complete evidence-to-claim mapping; organizing the evidence list before a hearing and clarifying the purpose of proof of each piece of evidence; reorganizing the chain of evidence to respond after the opposing party raises a rebuttal; a case involving multiple points of dispute, requiring evidence to be grouped and categorized under different argument claims; and assessing whether the existing evidence is sufficient to support all claims and identifying weak links and gaps in the chain of evidence. The core of this skill is establishing a complete mapping of "claim → element → evidence → probative value assessment", ensuring that every legal claim is fully supported by evidence, every piece of evidence has a clear purpose of proof, and the overall chain of argument is logically rigorous and clearly structured.
Pull the single best interaction from a video as a crisp looping GIF - delivered through a Clueso article/document, or as a short looping video clip if that fits the destination better - for changelogs, newsletters, and in-app tooltips. Use when the user says "make a GIF of this moment", "extract a GIF from this video", "I need a looping clip of the key interaction", "GIF for the changelog", or "grab the money shot as a GIF".
Use this skill to read device logs for test failures, debug, smoke testing, and CI repros. Covers `adb logcat` (stream), `adb logcat -d` (dump and exit), `adb logcat -c` (clear), buffer selection (`-b main|system|crash|events|radio|kernel|all`), priority ladder (V/D/I/W/E/F/S), filter expressions like `MyApp:D *:S`, format flags (`-v threadtime`, `-v json` on Android 11+), `--pid $(adb shell pidof -s pkg)`, time/count filters (`-T '01-01 12:00:00.000'`, `-t 100`), buffer rotation (`-r <kbytes>`, `-n <count>`, `-f <file>`), buffer sizing (`-G`, `-g`), and stripping `Log.d` calls in release builds via R8 `-assumenosideeffects`. If the user mentions "logcat filter only my app", "events buffer am_proc_start", "logcat json format", "grep logcat expensive", "missing logs after restart", "stripping Log.d release", or "logcat -f writes to host or device", use this skill.
Edit existing videos using AI — remix style, edit content, upscale resolution, remove background, or add audio/sound effects. Use when the user requests "Edit video", "Remix video", "Upscale video", "Remove video background", "Add sound to video", "Video to audio".
Draw **animated graphics / interactive charts / data dashboards** inside Feishu Docs — Miaobi BOX is the only carrier in Feishu Docs that can actually run CSS/JS (iframe sandbox). It can draw: the full ECharts suite (line/bar/pie/radar/scatter/heatmap/Sankey/funnel/gauge/candlestick/boxplot/parallel/sunburst/treemap/force-directed graph/timeline/Gantt), real maps with lat-long flying lines, echarts-gl 3D (map3D/3D bar/3D scatter/3D surface), real Three.js 3D scenes, word clouds, liquid fill, pure CSS animation, Canvas particles, SVG vector animation, scrolling KPI dashboard. When the user asks to "draw in Feishu Docs / make animations / animated graphics / interactive charts / data dashboard / Dashboard / line chart / bar chart / map / flying line chart / 3D chart / relationship graph / process animation / ECharts / visualization", or wants to build an **interactive document mini-app that can call AI / read and write Base tables / persist state / get user identity**, or mentions "Miaobi BOX / HTML widget / make the graphics in Feishu Docs move / embed a web page into Feishu Docs / window.magic", **you must use this skill**. Note: to be "animated" you can only use Miaobi BOX; the SVG nodes of the board (feishu-cli-board) are rasterized into static images by the server and will not move.
Generate images with Flux 2 Klein (Black Forest Labs' distilled fast variant of Flux 2) on RunComfy — bundled with the model's documented prompting patterns so the skill gets sharper output than naive prompting against the same model. Documents Flux 2 Klein's strengths (sub-second latency, multi-reference brand styling, declarative subject-first prompts), the step-count strategy (4–8 for fast iteration, ~25 for polish), the 9B vs 4B variant trade-off, and when to route to Flux 2 Pro / Seedream 5 / GPT Image 2 instead. Calls `runcomfy run blackforestlabs/flux-2-klein/9b/text-to-image` (or `/4b/`) through the local RunComfy CLI. Triggers on "flux 2 klein", "flux-2-klein", "flux klein", "BFL flux 2", or any explicit ask to generate with this model
Turns a community or forum text answer into a shareable how-to video. Source is a Q&A-style thread - someone asked a specific question, someone answered with text steps - and the output preserves that "direct answer to your exact question" framing rather than being smoothed into a generic tutorial. Use when the user says "turn this forum answer into a video", "make a how-to video from this community post", "this Q&A thread would make a good short video", pastes a forum or community thread and asks for a video, or points at a support-forum answer they want turned into something shareable.
Use when implementing translations/i18n in Frappe v14-v16 apps. Covers _() in Python, __() in JavaScript, CSV translation files, bench commands, string extraction rules, lazy translation _lt(), PO/MO files [v15+], RTL support, and custom app translations. Prevents common mistakes with f-strings, concatenation, and template literals that break string extraction. Keywords: translation, i18n, _(), __(), _lt(), CSV, PO, gettext, translate my app, multi-language, text not translated, wrong language, how to add translation. bench get-untranslated, RTL, localization.
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.
