Skills
98,067 packaged capabilities and playbooks for agents. Install any of them with one command.
Development
Inspect, verify, interact with, or profile a live Vite app by running temporary scripts in its browser page. Use for runtime state, physical input, browser logs, screenshots, recordings, and performance artifacts; prefer static inspection for questions the source can answer
Audit and improve JavaScript/TypeScript documentation including JSDoc comments (@param, @returns, @template, @example), comment markers (TODO, FIXME, HACK), and code comment quality. Use when asked to 'add JSDoc', 'document this function', 'audit documentation', 'fix comments', 'add TODO/FIXME markers', or 'improve code documentation'.
Answer questions about the AI SDK and help build AI-powered features. Use when developers: (1) Ask about AI SDK functions like generateText, streamText, ToolLoopAgent, embed, or tools, (2) Want to build AI agents, chatbots, RAG systems, or text generation features, (3) Have questions about AI providers (OpenAI, Anthropic, Google, etc.), streaming, tool calling, structured output, or embeddings, (4) Use React hooks like useChat or useCompletion. Triggers on: "AI SDK", "Vercel AI SDK", "generateText", "streamText", "add AI to my app", "build an agent", "tool calling", "structured output", "useChat".
Use for Blossom Carousel React or Next.js installation, stylesheet setup, JSX examples, Tailwind examples, accessibility or a11y guidance, navigation controls (BlossomPrev, BlossomNext, BlossomDots, BlossomDot), command events, overscroll events, and React-specific integration.
Decision guide for delegating to caveman-style subagents. Tells the main thread WHEN to spawn cavecrew-investigator (locate code), cavecrew-builder (1-2 file edit), or cavecrew-reviewer (diff review) instead of doing the work inline or using vanilla Explore. Subagent output is caveman-compressed so the tool-result injected back into main context is ~60% smaller — main context lasts longer across long sessions. Trigger: "delegate to subagent", "use cavecrew", "spawn investigator/builder/reviewer", "save context", "compressed agent output"
CesiumJS spatial math - Cartesian3, Cartographic, Matrix4, Quaternion, Transforms, Ellipsoid, BoundingSphere, projections, coordinate conversions. Use when converting between coordinate systems, computing positions on the ellipsoid, performing spatial intersection tests, building model matrices, or working with geographic projections.
Generates clean React/Vue component skeletons with TypeScript types, prop variants, styling hooks, test files, Storybook stories, and usage documentation. Use when users request "create a component", "scaffold component", "new React component", or "generate component boilerplate".
Write and revise queries with the published HelixDB Rust SDK (`helix-db = "3.0.0"`). Use for `read_batch`, `write_batch`, `#[query]`, direct `QueryRequest` values, traversal builders, projections, indexes, vector and BM25 search with traversal-scoped prefiltering, structured remote errors, and `Client::query`. Inspect local labels, edges, properties, and existing query patterns before inventing code. Stored routes, registration, and query bundles are not v3 SDK APIs. When the target is Helix Cloud, always use helix-mcp first.
Plan and run a media pipeline for images, video, and downloadable assets. Use this skill when designing image storage and delivery, choosing formats (WebP, AVIF), setting up responsive images, picking a video host, organizing a brand asset library, or auditing a slow image pipeline. Triggers on image pipeline, asset library, DAM, image optimization, WebP, AVIF, responsive images, video hosting, image CDN, asset workflow, media management. Also triggers when images are slow, broken, or scattered across systems.
Monitoring and observability patterns for Prometheus metrics, Grafana dashboards, Langfuse v4 LLM tracing (as_type, score_current_span, should_export_span, LangfuseMedia), and drift detection. Use when adding logging, metrics, distributed tracing, LLM cost tracking, or quality drift monitoring.
Extract formal spec and comprehensive test suites from existing MoonBit implementations. Use when asked to "extract spec from implementation", "generate tests from code", or "create spec-driven tests for existing package". Analyzes existing code to produce spec.mbt with `declare` keyword stubs and organized test files (valid/invalid).
Build new Next.js applications or migrate existing frontends (React, Vue, Angular, vanilla JS, etc.) to Next.js + shadcn/ui with systematic analysis and conversion. Enforces shadcn design principles - CSS variables for theming, standard UI components, no hardcoded values, consistent typography/colors. Use for creating Next.js apps, migrating frontends, adopting shadcn/ui, or standardizing component libraries. Includes MCP integration for shadcn documentation and automated codebase analysis.
Implements reliability patterns including circuit breakers, retries, fallbacks, bulkheads, and SLO definitions. Provides failure mode analysis and incident response plans. Use for "SRE", "reliability", "resilience", or "failure handling".
Produce stable, package-scoped Markdown documentation for many custom ABAP objects at once — classifies each by style (Classic/Modern/Mixed) and summarizes purpose and dependencies. Use when asked to "document this package", "generate onboarding docs for Z code", "seed a docs folder", or "knowledge transfer for this package"
Use when building or fixing game UI in Godot 4.7: main menu, settings screen, HUD (health bar, ammo, minimap frame), Control layout, containers, anchors, themes in code, 'UI scales wrong on phones', 'text too small on 4K', notch or safe area, stretch mode and aspect, gamepad or keyboard menu navigation, focus lost, input rebinding screen, localization (CSV, PO, tr_n plurals, Arabic RTL, CJK fonts, pseudolocalization), screen reader accessibility, UI scale slider, or clicks not reaching the game.
Use when the user wants to replace a whole Spree 6 service, workflow or API serializer with their own class — swap the add-to-cart workflow, checkout completion, cart totals recalculation, order status writer, storefront access policy, or an API v3 serializer. Common phrasings include "Spree.dependencies", "Spree::Dependencies", "swap a service", "replace Spree::Carts::AddItem", "cart_add_item_workflow", "carts_complete_workflow", "cart_recalculate_totals_workflow", "storefront_access_policy_class", "override an API serializer", "Spree.api.product_serializer", "dependency injection in Spree", "spree:dependencies:list", "spree:dependencies:overrides", "my cart_add_item_service override stopped working", "what can I swap". Covers when a swap is justified (usually a workflow hook is better), subclassing a shipped workflow, service vs workflow seams, legacy *_service keys that are stashed not applied, API serializer seams, per-controller overrides, and the introspection rake tasks.
Implement Syncfusion Angular AppBar component for navigation, branding, and actions. Use this skill when user needs to add top or bottom navigation bar, sticky headers, AppBar with buttons/menus/sidebars, positioning options, sizing modes (Regular/Prominent/Dense), color modes (Light/Dark/Primary/Inherit), and styling customization. Covers AppBar properties like isSticky, enablePersistence, enableRtl, htmlAttributes, responsive navigation patterns, action bars, header layouts, and navigation components.
Use when working with The Open Network (TON) blockchain, or when the user mentions TON-ecosystem terms the agent may not recognize—such as Tact, FunC, Tolk, Fift, TL-B, TVM, cells, BoC, Jettons, TEPs, TON Connect, workchains, shardchains, or liteservers. Provides a docs-first workflow for fetching and navigating TON documentation accurately, and covers smart contract development, transaction mechanics, wallet standards, token standards, Telegram Mini Apps, and infrastructure.
Use when you need Kafka messaging in Micronaut—including @KafkaClient and @KafkaListener design, @Serdeable serialization, topic/partition strategy, TestPropertyProvider integration tests, retries and dead-letter processing, and error handling. This should trigger for requests such as Add Kafka in Micronaut; Review Micronaut Kafka listeners; Improve retry and failure handling for Micronaut Kafka; Configure Micronaut Kafka clients listeners or SerDes; Add Micronaut Kafka retry or dead-letter handling. Part of Plinth Toolkit
Use this skill when auditing a codebase for AI agent readiness, or when guiding improvements to make a codebase work well with agentic coding tools. It applies when users ask to evaluate test coverage, file structure, type system usage, dev environment speed, or automated enforcement—the five pillars that determine how effectively coding agents can operate in a project. Triggers on “audit my codebase,” “make this agent-ready,” “improve for AI agents,” “agent-friendly,” or questions about why agents struggle with a codebase.
Generates new Capacitor plugin scaffolds and first-pass implementations from conversational requirements or a structured YAML input contract. Use when a user says "generate a Capacitor plugin", "create a Capacitor plugin scaffold", "build a native plugin for iOS and Android", "turn this plugin plan into Capacitor code", or "use this YAML contract to generate a plugin". Do not use for analyzing Cordova source, migrating whole apps, installing existing plugins, upgrading Capacitor versions, or publishing production-ready code without human review.
Reference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).
Debugs why session recordings aren't appearing in the local dev environment. Use when a developer reports that local replay ingestion isn't working, recordings aren't showing up despite /s calls, or the replay pipeline seems broken after hogli start. Covers the full local pipeline: SDK capture, Caddy proxy, capture-replay (Rust), Kafka, ingestion-sessionreplay (Node), recording-api (Node), SeaweedFS, and common failure modes like orphaned processes, stuck phrocs workers, and trigger misconfiguration.
You are a dependency security expert specializing in vulnerability scanning, license compliance, and supply chain security. Analyze project dependencies for known vulnerabilities, licensing issues, outdated packages, and provide actionable remediation strategies
Implements secure JWT (RFC 7519) signing and verification using HMAC-SHA256, RSA-PSS, ES256, and EdDSA, including token expiration, claims validation, and defenses against algorithm-confusion, none-algorithm, and key-injection attacks. Use when adding or hardening JWT-based authentication/authorization, or auditing token verification code for common JWT vulnerabilities
Write and improve programs that call Jev, TypeSafe's System One model. Use when designing TypeSafe questions (Choice, Score, Noul, Box), structuring state, sending images, composing answers in code, setting confidence thresholds, or diagnosing a Jev question that answers wrong or with low confidence.
Refactor MoonBit code to be idiomatic: shrink public APIs, convert functions to methods, use pattern matching with views, add loop invariants, and ensure test coverage without regressions. Use when updating MoonBit packages or refactoring MoonBit APIs, modules, or tests.
Context map of Prelude, the Nix devshell UI suite (MOTD banner, x command catalogue, docs viewer, themed prompt). Use for any task in a prelude-enabled repo, or when the user mentions prelude, x/motd/docs commands, the prelude wizard, or its flake-parts options. Enriches the task with accurate public-interface knowledge; never installs or reconfigures on its own.
Manage a consumer repo's Prelude docs viewer content — prelude.docs.pages, groups, mdSplit README hero, and generated nixosOptions pages. Use when adding or fixing project documentation shown by `docs` / `x docs`, wiring the README landing page, or generating option docs.
Build Retrieval-Augmented Generation (RAG) applications that combine LLM capabilities with external knowledge sources. Covers vector databases, embeddings, retrieval strategies, and response generation. Use when building document Q&A systems, knowledge base applications, enterprise search, or combining LLMs with custom data.
Use when a Godot game stutters, drops frames, hitches when entering new areas, overheats a phone, or must ship: profiling and Performance monitors, CPU or GPU bound, draw calls, MultiMesh, occlusion culling, LOD, shader compilation hitches, export to macOS, Windows, Linux, Steam, Android (APK, AAB, keystore), iOS (Xcode), web (itch.io, COOP/COEP headers, SharedArrayBuffer, JavaScript interop), build size budgets, demo and DLC builds, signing and notarization.
Use when designing, reviewing, or fixing WinUI 3: sample/control discovery with winapp find-ui, project-aware API checks with winapp find-api, layout planning, Fluent Design, theming, accessibility, and XAML binding. Requires WinApp CLI 0.7+. Load before authoring new XAML, reviewing UI PRs, migrating desktop UI, or choosing WinUI controls/patterns. Also use to find WinUI Gallery or Community Toolkit samples.
Build and review Wix CLI app extensions — dashboard pages, modals, plugins, menu plugins, custom element widgets, Editor React components, site plugins, embedded scripts, backend APIs, backend events, service plugins, data collections, and App Market readiness. Use when building any feature or extension for a Wix CLI app or preparing a Wix app for App Market review. Triggers on: add, build, create, implement, help me, dashboard, widget, plugin, backend, API, event, collection, embedded script, service plugin, Editor React component, checkout, shipping, tax, discount, SPI, CMS, schema, tracking, popup, admin panel, menu item, modal, validate, test, verify, register extension, App Market, app review, submission readiness.
Use when developer-owned execution needs inspectable branching, concurrency, joins, waits, approval, retry, runtime streams, pause/resume, recovery, or mixed sync/async orchestration. Use agently-stage instead when a provider-owned sync wrapper only bridges an async SDK
Benchmark similar product documentation and API documentation across Alibaba Cloud, AWS, Azure, GCP, Tencent Cloud, Volcano Engine, and Huawei Cloud. Given one product keyword, auto-discover latest official docs/API links, score quality consistently, and output detailed prioritized improvement recommendations.
Design, build, deploy, test, and debug serverless applications with AWS Lambda. Triggers on phrases like: Lambda function, event source, serverless application, API Gateway, EventBridge, Step Functions, serverless API, event-driven architecture, Lambda trigger. For deploying non-serverless apps to AWS, use deploy-on-aws plugin instead
Create Git branches following the Conventional Branch specification (feature/, bugfix/, hotfix/, release/, chore/, plus AI agent prefixes ai/, copilot/, cursor/, claude/, codex/). Use when creating a new branch, naming a branch, or checking whether a branch name complies with the spec
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.