Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Fix a reported bug in React Router from a GitHub issue. Use when the user provides a GitHub issue URL and asks to fix a bug, investigate an issue, or reproduce a problem. Handles the full workflow: fetching the issue, finding the reproduction, writing a failing test, and implementing the fix.
Operate and verify Flutter apps on a concrete emulator, simulator, browser, desktop, or physical device. Use for target lifecycle, platform-state checks, runtime evidence, and live integration flows; pair with flutter-navigation for route correctness, and do not select it for widget or golden-only work
Implement, review, or verify Flutter routing, deep-link destinations, back-stack behavior, nested navigation, restoration, redirects, and URL synchronization. Use for route correctness; pair with flutter-device-testing on a concrete runtime, and do not introduce a routing package by default.
Threat-model, audit, or harden Flutter and Dart applications against concrete security and privacy risks involving authorization, secrets, sensitive storage, cryptography, WebViews, deep links, platform exposure, or dependencies. Use for security findings and control requirements; route sign-in and session implementation to flutter-authentication.
Use whenever writing, debugging, or reviewing code that imports @formkit/core — FormKit's framework-agnostic core (node trees, values, events, hooks, message stores, ledgers, config, schema types, expression compiler). Provides correct 2.1.x API signatures, common-task examples, behavioral rules, and version limits.
Creates unique, production-grade frontend interfaces with exceptional design quality. Use when user asks to build web components, pages, materials, posters, or applications (e.g., websites, landing pages, dashboards, React components, HTML/CSS layouts, or styling/beautifying any web UI). Generates creative, polished code and UI designs that avoid mediocre AI aesthetics.
Use this skill whenever the user's main ask is on-chain, protocol, or Web3 behavior (not a pure safety verdict). Covers address and token on-chain analysis, platform metrics, reserves, heatmaps, stablecoins, bridges; optional news/UGC. Legacy alias: gate-info-defianalysis. Delegate: gate-info-risk (safety), gate-info-research (broad research), gate-news-intel (events/sentiment). v0.7.6+ shortcuts (+address-tracker, +token-onchain, +community-scan); legacy fallback below 0.7.6.
Fetch GitHub issues, spawn sub-agents to implement fixes and open PRs, then monitor and address PR review comments. Usage: /gh-issues [owner/repo] [--label bug] [--limit 5] [--milestone v1.0] [--assignee @me] [--fork user/repo] [--watch] [--interval 5] [--reviews-only] [--cron] [--dry-run] [--model glm-5] [--notify-channel -1002381931352]
Generates GTM implementation documentation, reporting impact analysis, GA4 report configurations, and stakeholder summaries. Use when users need to "document GTM implementation", "what reports can I build", "create event schema docs", "generate stakeholder summary", "analyze reporting impact", or want to understand business value of tracking data. Creates technical documentation, suggests GA4 explorations, defines remarketing audiences, and translates technical events into business insights.
Builds or changes a saved network enrichment workflow from connections or followers, with enriched people, their current companies, and links in both directions. Triggers on requests such as "enrich my network", "build a workflow for my followers", or "enrich these connections and their employers". Not for retrieving a list without enrichment, generic lead enrichment, one-off prospect qualification, or running an unchanged saved workflow, which belongs to gtm-workflow.
DWS cluster high memory root cause diagnosis skill, based on KooCLI v3.2.0+ and DWS Autopilot MCP Server. Automatically collects memory metrics, analyzes root causes (customer-side / system-side), and outputs a standardized diagnosis report. Applicable to scenarios where DWS cluster memory usage is too high, memory alarms, and OOM. Trigger words: 内存高, 内存告警, 内存诊断, 内存使用率过高, 内存不足, OOM, 内存溢出, 动态内存使用率超阈值, high memory, memory alarm, memory diagnosis.
Use when writing IAM policies, troubleshooting permission denied errors, setting up dynamic groups, or implementing IDCS integration. Covers policy syntax gotchas, principal types confusion, compartment inheritance, and common authorization failures.
Generates a CLAUDE.md file with AI-driven environment detection and advanced configuration options. This skill should be used when the user asks to "initialize config", "setup claude config", "create CLAUDE.md", or needs help configuring project instructions.
Score how ready a website, domain, or public MCP endpoint is for AI agents using Is Agentic, and read or act on the resulting report. Use when asked to check a site's agent readiness, get its Is Agentic score, fetch a scored report as JSON, or fix the issues that lower a site's score.
Provides comprehensive guidance for adding Java code comments following industry standards and best practices. This skill helps add class-level comments, method-level comments, and field-level comments to Java code. Use when the user wants to add comments to Java code, needs to document Java classes/methods/fields, wants to improve code documentation, or needs to generate JavaDoc comments. This skill covers Controller, Service, ServiceImpl, Mapper, Model, Entity, BO (Business Object), DTO, VO, and other common Java component types. The skill follows a systematic workflow: scan codebase, identify components, create todo list, and add comments in order (class comments → method comments → field comments).
Use the JetBrains IDE MCP Server (IntelliJ IDEA 2025.2+) for IDE-backed indexing, inspections, symbol info, project file operations, and structural refactoring. Use it when you need exact project-aware analysis or safe IDE refactors, rather than broad semantic repository discovery.
Build Solana DeFi applications combining Jupiter APIs with Helius infrastructure. Covers token swaps (Swap API V2), lending/borrowing (Lend protocol), limit orders (Trigger), DCA (Recurring), token/price data, transaction submission via Sender, fee optimization, real-time streaming, and wallet intelligence.
Run and interpret one safe, bounded Kitaru replay comparison against an accepted cohort and exact evaluators. Use when a user wants to replay a cohort, test or compare a model, prompt, system prompt, parameter, agent version, or tool policy, supervise an experiment run, determine whether one candidate helped, or ask for one bounded change worth testing.
Triggered when the user asks to explain, distinguish, break down, or understand a legal concept. Typical trigger scenarios include but are not limited to: - The user directly asks "what is XX" or "how to understand XX" - The user asks to compare two legal concepts (e.g. "the difference between unjust enrichment and negotiorum gestio") - The user provides specific facts and asks whether they constitute a legal concept (e.g. "A mistakenly transferred money to B; does this constitute unjust enrichment?") - The user encounters unfamiliar terminology in legal analysis, document drafting, or contract review - The user asks to break down the elements of a concept or analyze its legal effects. This skill is the basic unit of legal analysis, providing a conceptual foundation for applying legal rules, case analysis, legal document drafting, and more.
Trigger this skill when the following situations arise in legal reasoning: (1) the text of a legal provision is ambiguous and open to multiple interpretations; (2) disputes arise when applying a provision to specific case facts; (3) it is necessary to argue that a certain provision should or should not apply to a particular situation; (4) the two parties disagree on the interpretation of the same provision; (5) provisions appear to conflict and need to be reconciled through interpretation; (6) novel case facts cannot be directly mapped to the text of an existing provision; (7) it is necessary to analyze or challenge the application of law in a judicial document. The core of this skill is to use legal interpretation methods such as literal, systematic, and teleological interpretation to conduct rigorous interpretive argumentation on key provisions, arrive at reasonable and defensible legal meanings, and present the interpretation process and conclusions in a standardized manner.
Look up public product details on TikTok Shop. Enter a full product URL or a 19-digit product ID, optionally specifying major sites in the US, UK, Southeast Asia and Europe, and it returns grouped data such as product title, category, price, sales, SKU inventory, images, shop, review overview, logistics and promotions. Triggered when the user mentions TikTok Shop product details, TikTok product link parsing, product ID lookup, price and inventory verification, SKU variants, competitor page breakdown, TikTok product details, or TikTok Shop listing lookup. Even if the user does not explicitly say "product details", this skill should also trigger when they want to read the current public product page data from a TikTok Shop product URL or a 19-digit product ID.
Operate, configure, secure, and troubleshoot the LiteLLM AI gateway (proxy) and Python SDK: run the proxy (litellm --config), route to 100+ providers through one OpenAI-compatible API, configure model lists and routing/reliability, virtual keys, teams, budgets, rate limits, caching, guardrails, observability, and spend, and diagnose request failures. Use when deploying or running a LiteLLM proxy or gateway (config.yaml, ghcr.io/berriai/litellm), wiring the Python SDK or OpenAI SDK through it, or hardening a public-facing deployment. Do not use for operating a single inference engine (vllm, llama-cpp), for engine-selection methodology (ml-engineering), or for building applications on top of an LLM API (backend/frontend engineering).
Operate, configure, benchmark, and troubleshoot llama.cpp across CPU, Metal, CUDA, HIP/ROCm, Vulkan, SYCL, and hybrid or multi-GPU systems. Use when installing or building llama.cpp, selecting or inspecting GGUF models, running llama-cli or bounded llama-completion, serving an OpenAI-compatible API with llama-server, tuning memory and performance, or diagnosing backend, context, template, and API failures. Do not use for model training or fine-tuning, general inference-framework selection, llama-cpp-python or other bindings, LlamaIndex, Ollama, or LM Studio operation.
Use when creating, applying, or validating overlay files including x-speakeasy extensions. Covers overlay syntax, JSONPath targeting, retries, pagination, naming, grouping, open enums, global headers, custom security. Triggers on "create overlay", "apply overlay", "overlay file", "x-speakeasy", "add extension", "configure retries", "add pagination", "overlay for retries".
MapStruct best-practice patterns. Distilled from the official Reference Guide, covering the three ways to define a Mapper (interface/abstract class/default method), Spring injection strategies (CONSTRUCTOR recommended / SETTER to resolve circular dependencies), nested mapping patterns (explicit sub-methods > one-off dot notation configuration), the safer choice of @Qualifier over @Named, Lombok integration rules, and cautious use of @MappingComposition. Corrects the most common LLM mistakes: using reflective BeanUtils instead of compile-time MapStruct, putting all nested mappings in a single method, and not knowing that SETTER resolves circular dependencies.
Set up and run Airtable-based marketing operations workflows — request intake, campaign orchestration, creative production, content calendars, brand and compliance review, events, localization, budgets and ROI, capacity planning. Use when the user wants a marketing request "front door," to manage campaigns, coordinate briefs and assets, build a content calendar, plan launches or events, track budgets, measure ROI, or set up agency multi-client delivery. Adapts to org size (solo marketer to enterprise multi-brand or agency) and integrates with or displaces tools like HubSpot, Marketo, Mailchimp, Klaviyo, Workfront, Asana, Monday, and Wrike. Asks scope first.
Author and render/verify mermaid diagrams (flowchart, sequence, state, class, ER, C4, gantt) using mermaid-cli (mmdc). Use whenever the user wants to create, update, verify, or render a mermaid diagram, or wants an architecture/flow/sequence/state/ER diagram in Markdown, or when a ```mermaid block needs checking for validity before it lands in a file. Covers the turnkey `mermaid-render` wrapper, the required headless-Chromium no-sandbox config, and common mermaid authoring pitfalls.
Enterprise middleware testing patterns for message routing, transformation, DLQ, protocol mediation, ESB error handling, and EIP patterns. Use when testing middleware layers, message brokers, ESBs, or integration buses.
Motoko actor state migration and schema evolution using the mops-managed migration chain (a migrations/ directory of timestamped files). Use when upgrading canisters, changing actor field types, renaming or adding stable fields, deciding implicit vs explicit migration, or writing migration files — including one-shot changes. For unexplained compatibility diagnostics, frozen migration files, a non-empty first OldActor, projects converted from legacy persistence, or requests to remove the chain or revert to inline (with migration = ...), use troubleshooting-motoko-migrations.
Operate Milvus vector database with pymilvus Python SDK. Use when the user wants to connect to Milvus, create collections, insert vectors, perform similarity search, hybrid search, full-text search, manage indexes, partitions, databases, or RBAC via Python code.
Guide users from zero to their first working mirrord session — check system requirements, install mirrord via CLI, VS Code, or IntelliJ, run a first session against a Kubernetes target, and verify the connection. Use when a user is new to mirrord, wants to install it, needs help running their first session, or wants to debug or test a local process against their Kubernetes cluster.
Use when the user asks to build Android screens, Jetpack Compose layouts, Material components, or any Android app UI. Also use when styling, redesigning, or beautifying any Android interface. Covers Material 3 Expressive for Android 16, spring motion, shape variety, and platform-native patterns.
Best-practice guide for authoring a Muxy extension — how it should look and behave so it reads as a native part of the app. Covers theming (follow the theme, never hardcode colors), the sizing scale, and which surface to use. Mechanics (manifest fields, permissions, the window.muxy API) live in the linked docs.
Analyze a Materialize environment via the MCP Developer endpoint, and/or configure an MCP client (Claude Code, Cursor, VS Code, Zed, Continue, Windsurf, Claude Desktop) to connect to the materialize-developer server. For analysis: check environment health, investigate performance, troubleshoot stale materialized views, diagnose memory pressure, find paused clusters holding back compaction, audit resource utilization, run EXPLAIN ANALYZE on user objects, and get optimization recommendations. For client connection: configure/connect/set up an MCP client to materialize-developer (Emulator, Cloud, or self-managed), control which user/role is used, and switch between identities. Trigger even if the user just says “check my environment,” “why is my MV stale,” “why is my cluster slow,” “what can I optimize,” “explain analyze my materialized view,” “how do I connect Claude Code to materialize-developer,” or “configure Cursor for the Materialize MCP.”
Guide to review incoming one-way (unilateral) commercial NDAs in a jurisdiction-agnostic way, from either a Recipient or Discloser perspective (user-selected), producing a clause-by-clause issue log with preferred redlines, fallbacks, rationales, owners, and deadlines.
Entry point for OmniRoute — local/remote AI gateway with OpenAI-compatible REST for chat, image, TTS, STT, embeddings, web search, web fetch, MCP, A2A. Use when the user mentions OmniRoute, OMNIROUTE_URL, or wants AI without writing provider boilerplate. This skill covers setup + indexes capability skills; fetch the relevant capability SKILL.md from the URLs below when needed.
Interactive OpenHarmony source code download with mirror selection (GitCode/Gitee/GitHub), environment checking, branch selection, and real-time progress. Use when user requests:"下载 OpenHarmony", "download OpenHarmony", "下载源码", "获取源码", "拉取代码", "clone openharmony", or "repo init".
Use the public `orca` CLI to operate Orca-managed worktrees, folder contexts, terminals, repos, automations, artifacts, worktree comments, and the browser embedded inside the Orca app. Use when the user says "$orca-cli", "use orca cli", "Orca worktree", "child worktree", "cardStatus", "spawn codex/claude in a worktree", "read/wait/send Orca terminal", "terminal send", "full handoff", "handover", "give this to another agent", "another worktree", "Orca browser", "orca artifacts", "share HTML/Markdown", "public artifact link", or "control the browser inside Orca". Prefer this over raw `git worktree`, ad hoc PTYs, Playwright, or Computer Use when the task touches Orca-managed state. Use Computer Use for browser windows, webviews, or desktop UI outside Orca's embedded browser.
Search papers across arXiv, DBLP, OpenAlex, OpenReview, Semantic Scholar, and Crossref for a given query and year range, using ./scripts/search_papers.py. Use when the user asks to find papers, related work, prior art, or recent publications on a specific topic, especially when they mention a date range or specific venues like NeurIPS, ICLR, or ICML.
Comprehensive guide for creating professional UI/UX designs in Penpot using MCP tools. Use this skill when: (1) Creating new UI/UX designs for web, mobile, or desktop applications, (2) Building design systems with components and tokens, (3) Designing dashboards, forms, navigation, or landing pages, (4) Applying accessibility standards and best practices, (5) Following platform guidelines (iOS, Android, Material Design), (6) Reviewing or improving existing Penpot designs for usability. Triggers: "design a UI", "create interface", "build layout", "design dashboard", "create form", "design landing page", "make it accessible", "design system", "component library".
Expert in observing, benchmarking, and optimizing AI agents. Specializes in token usage tracking, latency analysis, and quality evaluation metrics. Use when optimizing agent costs, measuring performance, or implementing evals. Triggers include "agent performance", "token usage", "latency optimization", "eval", "agent metrics", "cost optimization", "agent benchmarking".
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.