Skills
96,770 packaged capabilities and playbooks for agents. Install any of them with one command.
Development
glab CLI documentation — GitLab's official command-line tool. Covers merge requests, issues, CI/CD pipelines, repositories, runners, releases, clusters, stacks, variables, schedules, labels, milestones, auth, config, snippets, deploy keys, SSH/GPG keys, Duo AI, MCP, and OpenTofu. USE THIS SKILL WHEN the user asks about glab commands, glab CLI reference, glab mr, glab ci, glab issue, glab repo, glab runner, or GitLab CLI docs.
Design and review gRPC and Protocol Buffer APIs against Google API Improvement Proposals (AIPs). Use when adding or changing .proto resources, RPCs, standard or custom methods, pagination, filtering, field masks, long-running operations, soft delete, versioning, or when an AIP number or compliance review is mentioned. Do not use for protobuf syntax, code generation, or implementation unless AIP guidance is the task.
Covers the Google Gemini embeddings API (gemini-embedding-001) for RAG, semantic search, document clustering, and similarity matching: @google/genai SDK usage, fetch-based calls from Cloudflare Workers, batch embedding, the 8 task types (RETRIEVAL_QUERY, RETRIEVAL_DOCUMENT, SEMANTIC_SIMILARITY and others), dimension selection from 128 to 3072, cosine similarity, chunking, caching, and Cloudflare Vectorize integration. Prevents 8+ embedding errors including dimension mismatches, wrong task type, free-tier rate limiting (100 RPM), vector normalization mistakes, 2,048 token text truncation, and model version confusion. Use when generating text embeddings with Gemini, building a RAG or vector search pipeline, storing vectors in Vectorize, or debugging embedding dimension, rate limit, or normalization problems.
Google Cloud Pub/Sub messaging service. Covers topics, subscriptions, push/pull delivery, and dead-letter handling. Use for GCP-native event-driven architectures and real-time analytics. USE WHEN: user mentions "google pub/sub", "pubsub", "gcp messaging", "push subscription", "pull subscription", asks about "cloud pub/sub", "ordering keys", "bigquery subscription" DO NOT USE FOR: AWS-native - use `sqs`; Azure-native - use `azure-service-bus`; event streaming - use `kafka` or Dataflow; on-premise - use `rabbitmq` or `activemq`; multi-cloud - use `kafka` or `pulsar`
Render and view webpages using a headless browser (Playwright) to fetch JS-rendered HTML, extract visible text, and optionally save full-page screenshots. Use when a user asks to “无头浏览器打开/查看网页”, needs the rendered DOM instead of raw curl HTML, or wants a screenshot of a page.
Write HLD, High-Level Design, 写技术设计文档. Use when: system architecture design, technology selection, and technical solution definition are needed after the PRD and API Contract are complete. Also used for limited incremental updates to existing related documents.
Zoom out to see how code, features, or plans fit into the bigger picture. Maps bounded contexts, traces cross-cutting concerns, surfaces hidden dependencies, and identifies systemic risks. Use when you're lost in details, unfamiliar with an area, or need to understand the full impact of a change.
HTTP clients for frontend and Node.js. Covers Axios, Fetch API, ky, and ofetch. Includes interceptors, error handling, retry logic, and auth token management. Use for configuring API clients and HTTP communication. USE WHEN: user mentions "HTTP client", "Fetch API", "ky", "ofetch", "HTTP wrapper", "retry logic", "token refresh", asks about "which HTTP client to use", "HTTP request library", "API client setup", "request interceptors" DO NOT USE FOR: Axios-specific questions - use `axios` instead; GraphQL - use `graphql-codegen` instead; tRPC - use `trpc` instead; WebSocket connections
IntelliJ IDEA MCP for Java code analysis, testing, refactoring, and project navigation. Use when: (1) Checking Java compilation errors (get_file_problems is more accurate than LSP for multi-module Maven), (2) Performing rename refactoring across the codebase, (3) Searching code or finding files in the project. Requires IDEA to be running with MCP plugin connected.
Comprehensive frontend design skill — 23 commands for shaping, auditing, polishing, and refining UI. Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, empty states. Handles UX review, visual hierarchy, info architecture, cognitive load, accessibility, performance, responsive behavior, theming, anti-patterns, typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy, error states, edge cases, i18n, and reusable design systems/tokens. Also use for bland designs needing boldness, loud designs needing quiet, live browser iteration, or ambitious visual effects. Not for backend-only or non-UI tasks.
Build a whole feature by breaking a plan into tasks and running implement-and-review on each. Given a plan or feature, decompose it into tracer-bullet vertical-slice tasks (via the to-tasks skill — acceptance contracts + gate flags + dependencies), then execute the task DAG (independent tasks in parallel) by running implement-and-review for each per-task FE/BE cross-reviewed TDD build, integrate tasks in dependency order with per-task acceptance, and finish with a feature-wide full-review, leaving tests/build green. Opus orchestrates breakdown, scheduling, integration, and final review; per-task building is implement-and-review's job. Use to ship/build a multi-task feature end-to-end. For a single scoped task call implement-and-review directly. Distinct from models-consensus (deliberation only, produces no code) and from ship — ship is the idea→PR conductor, including the interactive spec/plan phases and PR delivery, while implement-feature is the plan→built-feature DAG executor with no spec phase and no PR.
Find deepening opportunities in a codebase, referencing the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly coupled modules, or make a codebase more testable and AI-navigable.
Transforms a rough, unpolished prompt idea into a precise, structured AI coding prompt. Automatically researches the current project context (stack, file structure, conventions, git history) before generating. This skill should be used when the user provides a vague or "dirty" prompt idea and asks to refine, improve, or rewrite it — e.g. "improve this prompt", "refine my prompt", or simply describes what they want done in rough terms. NOT when the user wants the work done or planned rather than the prompt itself — deep-plan produces the plan; just doing the task needs no prompt skill.
Create and use Inngest middleware for cross-cutting concerns. Covers the middleware lifecycle, creating custom middleware, dependency injection with dependencyInjectionMiddleware, encryption via @inngest/middleware-encryption, Sentry error tracking via @inngest/middleware-sentry, and custom middleware patterns.
Conduct iterative interviews to clarify scope, surface constraints, and reach a "do / don't do / done" agreement before starting non-trivial work. Use when receiving an ambiguous request, a vague task description, or before entering plan mode for any non-trivial implementation. Skip for one-line concrete fixes ("rename foo to bar"), single-fact questions, or when scope, constraints, and acceptance criteria are already specified. Supports requirements definition, debugging investigation, architecture review, security review, documentation creation, and general exploration.
Enforces JSDoc documentation standards for this TypeScript project. This skill should be used when writing or reviewing TypeScript code to ensure proper documentation with file preambles, function docs, interface docs, and the critical distinction between documenting "what" vs "why". Use this skill to understand the project's JSDoc ESLint rules and established patterns.
Debug LangChain and LangGraph agents by fetching execution traces from LangSmith Studio. Use when debugging agent behavior, investigating errors, analyzing tool calls, checking memory operations, or examining agent performance. Automatically fetches recent traces and analyzes execution patterns. Requires langsmith-fetch CLI installed.
This skill should be used when analyzing a downstream project's git diff after Lisa was applied to identify improvements that should be upstreamed back to Lisa templates. It validates the environment, captures the diff, correlates changes with Lisa template directories, categorizes each change, and offers to upstream improvements.
Apply Litestar dependency injection with Provide, Dependency markers, layered scopes, overrides, generator-based cleanup, per-request caching, and nested providers. Use when wiring services, settings, database sessions, request-derived resources, or test doubles into Litestar handlers, controllers, routers, and apps. Do not use when plain local function parameters or a small helper are sufficient.
LLM gateway and routing configuration using OpenRouter and LiteLLM. Invoke when: - Setting up multi-model access (OpenRouter, LiteLLM) - Configuring model fallbacks and reliability - Implementing cost-based or latency-based routing - A/B testing different models - Self-hosting an LLM proxy Keywords: openrouter, litellm, llm gateway, model routing, fallback, A/B testing
Write a local development environment setup guide for a service or project — covering prerequisites, repository setup, environment variables, local service dependencies, database seeding, running the service, running tests, common gotchas, IDE recommendations, and first-contribution checklist. Use when asked to write a dev setup guide, create onboarding documentation for engineers, document local environment setup, or write a getting-started guide for a codebase. Produces a complete setup guide that a new engineer can follow from zero to running tests in under 30 minutes, with a troubleshooting section for the most common setup failures.
Build production-ready web authentication using Lucia Auth and The Copenhagen Book patterns. Use when users ask to "add auth", "implement authentication", "setup login", "add sessions", "password reset", "verify email", "implement OAuth", "add 2FA/MFA", "add passkeys", "password hashing", or reference "lucia auth" or "the copenhagen book". Default to a production baseline for browser-based web apps, not a demo-only login flow.
MATLAB and GNU Octave numerical computing for matrix operations, data analysis, visualization, and scientific computing. Use when writing MATLAB/Octave scripts for linear algebra, signal processing, image processing, differential equations, optimization, statistics, or creating scientific visualizations. Also use when the user needs help with MATLAB syntax, functions, or wants to convert between MATLAB and Python code. Scripts can be executed with MATLAB or the open-source GNU Octave interpreter.
MatrixScan AR (Barcode AR, BarcodeAr) in .NET for Android projects (`net*-android` TFM, `Scandit.DataCapture.Barcode` NuGet, non-MAUI — MAUI apps use matrixscan-ar-net-maui) — scanning multiple barcodes at once with AR highlights and annotations. Use for integration, settings, listeners/events, highlight and annotation providers, lifecycle, SDK version migration, or troubleshooting.
MatrixScan AR (Barcode AR, BarcodeAr) in .NET for iOS projects (`net*-ios` TFM, `Scandit.DataCapture.Barcode` NuGet, non-MAUI — MAUI apps use matrixscan-ar-net-maui) — scanning multiple barcodes at once with AR highlights and annotations (info annotations, popovers, status icons) in C#. Use for integration, settings, listeners/events, highlight and annotation providers, torch/zoom/macro controls, lifecycle, SDK version migration, or troubleshooting.
Query the local Maven/Gradle artifact index from the terminal. Use when you see an import but cannot find the class definition in the workspace (it likely comes from a compiled internal library), when you need to read the actual implementation of an internal or non-well-known library, or when you need to find which artifact contains a class, discover implementations of an interface, or inspect proto/resource files inside JARs. Do not use for well-known standard libraries (java.util, Spring core, etc.) that the AI already knows.
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.