Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Use this skill when profiling or optimizing a PixiJS v8 app for FPS, draw calls, or GPU memory. Covers destroy patterns (cacheAsTexture(false), releaseGlobalResources), GCSystem and TextureGCSystem, PrepareSystem, object pooling, batching rules, BitmapText for dynamic text, culling (Culler, CullerPlugin, cullable, cullArea), resolution/antialias tradeoffs, low-level wins (transient MSAA render textures, render bundles, partial buffer updates). Triggers on FPS, jank, draw calls, batching, object pool, GCSystem, PrepareSystem, Culler, cacheAsTexture, memory leak, destroy patterns, render bundle, transient, MSAA, Buffer.update, WebGPU performance, TexturePool, repeatEdgePixels, bind group.
Live SEO data via DataForSEO MCP server: SERP analysis, keyword research (volume, difficulty, intent, trends), backlink profiles, on-page analysis, competitor and content analysis, business listings, AI visibility (LLM mention tracking), and domain analytics. Requires DataForSEO extension installed. Use when user says "dataforseo", "live SERP", "keyword volume", "backlink data", "AI visibility check", or "real search data".
Configure EC2 instances to securely call AWS services by creating and attaching IAM roles via instance profiles, eliminating hardcoded credentials. Use when an EC2 instance needs permission to access AWS services such as S3, DynamoDB, SQS, or CloudWatch through temporary credentials.
Run a comprehensive SEO audit on a website covering technical health, on-page optimization, content quality, and backlink profile. Use when the user asks for an SEO audit, site review, SEO health check, "what's wrong with my SEO", website analysis, or a full diagnostic of their site's search performance. For speed-specific issues, see audit-speed. For technical crawl/index issues only, see diagnose-seo.
Apply when designing types, reviewing a function signature, or writing code in any statically-typed language. Make illegal states unrepresentable, brand semantic primitives, parse external data at boundaries, exhaust variants, derive from authoritative schemas.
Given a domain, find the few independent forces that truly hold it up. A generator that reduces a dozen phenomena to the fewest possible terms — and can regenerate each phenomenon one by one — only then does it count. Use when user says '降秩', '找秩', '秩是什么', '这个领域靠什么撑着', '背后是什么', or wants to decompose any domain to its irreducible generators.
Use when working with Lynx DevTool or debugging a Lynx app, page, or device, especially when the task mentions clients or sessions, CDP or App commands, DOM/CSS inspection, runtime or console logs, evaluating JavaScript on a device, screenshots, heap snapshots, performance traces, Page.reload or App.openPage, global switches, interactive snapshot refs and taps/fills/scrolls, inspecting or searching a ReactLynx component tree, linking DOM Snapshot refs to ReactLynx components (`reactlynx link`), or mutating ReactLynx props/state/context on Android, iOS, or Desktop.
Query Ondo tokenized US stock data on Binance Web3. Covers: supported stock token list, RWA metadata (company info, attestation reports), market and per-asset trading status (with corporate action codes for earnings, dividends, splits), real-time on-chain data (token price, holders, circulating supply, market cap), US stock fundamentals (P/E, dividend yield, 52-week range), and token K-line/candlestick charts. Use this skill when users ask about: - Tokenized stock price, holders, or on-chain data for specific tickers - Whether a stock token is tradable, paused, or halted - Ondo RWA token list or which US stocks are available on-chain - Corporate actions affecting a token (dividends, stock splits, earnings halt) - Stock token K-line or candlestick chart data - Comparing on-chain token price vs US stock price NOT for general crypto tokens (BTC, ETH, SOL, etc.) — use query-token-info for those.
Design an auditable playbook when no narrower one fits: a large migration, an ambitious multi-part change, or work a human reviews after stepping away. Scale rigor to the task, run a hypothesis loop, and log decisions. Use for /figure-it-out or "figure it out".
Keep a reviewable decision trail for long-running or unattended work: a TSV log with one row per decision (what, why, evidence, result). Local by default; commit it when a reviewer needs the trail to trust the result. Use for /show-me-your-work, autonomous or multi-phase runs, or work a human reviews after stepping away.
Use for any task involving Dagster, the dg CLI, assets, materialization, components, data tools or data pipelines, matching the situations the dagster skill handles. This is the deprecated dagster-expert plugin. It carries no Dagster guidance and exists only to tell the user to move to the dagster plugin.
Turns an answer into a one-page visual HTML explainer: the model writes a short Markdown draft and the bundled CLI builds one page. Use it proactively and liberally, without being asked, whenever a page would help the reader more than plain text, even if plain text would also work: any explanation of how something works or how parts relate (flow, request path, architecture, code or folder structure, state machine, lifecycle, history); any comparison, trade-off or decision with pros and cons; any diagnosis, review or investigation, especially with several causes or findings to rank; any answer with a table, a numbered or ranked list, steps, branches or several sections; or when the user says "I don't get it / draw it / explain visually / 讲讲原理 / 没看懂 / 画个图". When in doubt, use it: a page is quick and cheap to make. Also for explainer videos ("make a video", "做个视频") and for changing its settings. Skip only for small talk, a trivial one-line answer, or when the user asks for plain text.
Use when the user asks how to build with OpenAI products or APIs, asks about Codex itself or choosing Codex surfaces, needs up-to-date official documentation with citations, help choosing the latest model for a use case, or model upgrade and prompt-upgrade guidance; use OpenAI docs MCP tools for non-Codex docs questions, use the Codex manual helper first for broad Codex self-knowledge, and restrict fallback browsing to official OpenAI domains.
Use MagicPath through the magicpath-ai CLI to find, preview, inspect, install, export, create, and edit UI components and manage MagicPath skills. Trigger for MagicPath designs/components; personal or team projects; active canvases and selected components/images/revisions; themes/design systems; teams, members, ownership, attribution, installed-component audits, and share/view links. Use when exporting exact MagicPath components or revisions to local folders/apps; replacing, adapting, or translating local UI with 1:1 fidelity and explicit requested changes; installing React/TypeScript components with inspect/add; authoring responsive interactive canvas components with code start/submit; importing or recreating UI from local paths or GitHub/GitLab/Bitbucket repositories into MagicPath; or creating, retrieving, updating, importing, enabling, disabling, or deleting MagicPath skills. In hosts with an embedded browser, keep the project canvas open via share URLs for visual work.
Builds and customizes assistant-ui chat UI from composable, unstyled @assistant-ui/react primitives: ThreadPrimitive, ComposerPrimitive (with Unstable_TriggerPopover for mentions and slash commands), MessagePrimitive and MessagePartPrimitive, ActionBarPrimitive, BranchPickerPrimitive, AttachmentPrimitive, ErrorPrimitive, AssistantModalPrimitive, ChainOfThoughtPrimitive, SelectionToolbarPrimitive, SuggestionPrimitive, QueueItemPrimitive, and the ThreadList primitives. Use when assembling or styling a custom Thread, Composer, message list, action bar, branch picker, or suggestion grid; wiring composer input, dictation, quoting, or attachments; or handling editing, branching, virtualization, scrolling, or chain of thought UI. Covers MessagePrimitive.Parts render functions per part type, groupPartByType, AuiIf in selector form, and the unstyled gotchas (AssistantRuntimeProvider, className). For prebuilt UI use setup, for the styled catalog use elements, for multi-thread sidebars use thread-list
Creates a production-ready VPC with public and private subnets across multiple Availability Zones, including internet gateway, NAT gateways, route tables, and security groups following AWS Well-Architected principles. Use when deploying multi-AZ VPC infrastructure with automatic CIDR planning and DNS resolution.
Design habit-forming product loops using the Hook Model (Trigger, Action, Variable Reward, Investment). Use when the user mentions "users arent coming back", "habit formation", "engagement loops", "habit zone", or "the manipulation matrix". Also trigger when designing notification or re-engagement strategies, building streaks or progress systems, or analyzing why users stop after signup. Covers ethics evaluation and onboarding for habits. For friction reduction and B=MAP, see improve-retention. For viral sharing, see contagious.
Interactive PDF viewer. Use when the user wants to open, show, or view a PDF and collaborate on it visually — annotate, highlight, stamp, fill form fields, place signature/initials, or review markup together. Not for summarization or text extraction (use native Read instead).
Provides instructions for integrating the Google Mobile Ads (GMA) SDK. Use this skill when the user wants to get started with, install, integrate, set up, or configure the SDK for AdMob or Ad Manager, GMA Next-Gen SDK or mobile ads framework in an Android, iOS, or Unity application.
Use this skill when understanding how PixiJS v8 renders frames: the systems-and-pipes renderer, the render loop, and how the library adapts to different environments. Covers WebGLRenderer/WebGPURenderer/CanvasRenderer selection, renderer.render() pipeline, environment detection, rendering into RenderTextures and RenderTargets (flipY, mipLevel, layer, bind/push/pop, copyDepthTexture), WebGL context loss and WebGPU device loss recovery, and pointers to per-topic deep dives. Triggers on: renderer, WebGL, WebGPU, Canvas, render loop, render pipeline, systems, environments, autoDetectRenderer, RenderTexture, RenderTarget, render to texture, flipY, mipLevel, layer, 3D texture, depth slice, renderTarget.bind, copyToTexture, copyDepthTexture, depth-only, WebGLLoader, WebGPULoader, CanvasLoader, RendererLoader, lazy-load renderer systems, code splitting, context lost, webglcontextlost, device lost, GPUDevice.lost.
Delegate a coding task to a separate Claude Code CLI process or another Claude session as implementer, then review its diff and land it yourself. Use only when the user explicitly asks to delegate implementation to Claude Code, another Claude session, or the `claude` CLI — for example, "have another Claude implement this", "delegate this to Claude Code", or "run this queue through a separate Claude session." Do not trigger merely because the current orchestrator is Claude, and do not use when the user asks the current Claude to implement directly without delegation.
Use this skill to 'check my DeFi positions', 'view DeFi holdings', 'show my DeFi portfolio', 'what DeFi am I invested in', 'show my staking positions', 'show my lending positions', 'DeFi balance', 'DeFi 持仓', '查看DeFi持仓', '我的DeFi资产', '持仓详情', '持仓列表', or mentions viewing DeFi holdings, positions, portfolio across protocols — when the user does NOT name a specific DApp. Covers positions overview and per-protocol position detail. Do NOT use for deposit/redeem/claim operations — use okx-defi-invest. Do NOT use for wallet token balances — use okx-agentic-wallet. Do NOT use for DEX spot swaps — use okx-agentic-wallet. Do NOT use when the user names a specific third-party DApp by name (e.g. 'show my Aave positions', 'my Hyperliquid balance', 'check my Polymarket holdings') — route to okx-dapp-discovery instead, which loads the DApp's plugin for protocol-native position views.
Guide to the assistant-ui runtime system, the state and action layer behind every chat surface. Use when creating a runtime (useLocalRuntime with a ChatModelAdapter, useExternalStoreRuntime for Redux- or Zustand-backed stores, useRemoteThreadListRuntime, useCloudThreadListRuntime, useAssistantTransportRuntime), wiring AssistantRuntimeProvider, or reading and mutating thread, message, composer, and attachment state. Covers useAui, useAuiState, and useAuiEvent with the stable selector rule, the scope accessors (aui.thread, aui.message, aui.composer, and siblings), the AuiConfig plus AuiProvider grammar, RuntimeCapabilities, and the attachment, speech, dictation, feedback, suggestion, and history adapters. Refer here for a provider error ('Cannot read property of undefined'), a removed v0.12 hook import, an infinite re-render from an object literal selector, or state not updating. For multi-thread list UI use thread-list instead; for toolkit definitions and tool call UI use tools instead
Use this skill when drawing vector shapes and paths in PixiJS v8. Covers the Graphics API: shape-then-fill methods (rect/circle/ellipse/poly/roundRect/star/regularPoly/roundPoly/roundShape/filletRect/chamferRect), path methods (moveTo/lineTo/bezierCurveTo/quadraticCurveTo/arc/arcTo/arcToSvg/closePath), fill/stroke/cut, holes, FillGradient (linear/radial), FillPattern, GraphicsContext sharing, svg import/export, containsPoint hit testing, cloning, clearing, bounds, fillStyle/strokeStyle, draw-time transforms (rotateTransform/scaleTransform/translateTransform/setTransform/save/restore), default styles, GraphicsPath reuse. Triggers on: Graphics, GraphicsContext, rect, circle, poly, roundRect, fill, stroke, cut, hole, beginHole, FillGradient, FillPattern, moveTo, bezierCurveTo, svg, graphicsContextToSvg, svg export, GraphicsOptions, containsPoint, clone, clear, bounds, rotateTransform, translateTransform, setFillStyle, setStrokeStyle, GraphicsPath
Use when investigating or improving WordPress performance (backend-only agent): profiling and measurement (WP-CLI profile/doctor, Server-Timing, Query Monitor via REST headers), database/query optimization, autoloaded options, object caching, cron, HTTP API calls, and safe verification.
Design scalable distributed systems using structured approaches for load balancing, caching, database scaling, and message queues. Use when the user mentions "system design", "scale this", "high availability", "rate limiter", "design a URL shortener", "design Twitter", "design Uber", "design a news feed", "system design interview", "capacity planning", or "distributed architecture". Also trigger when estimating infrastructure requirements, choosing between microservices and monoliths, or designing for millions of concurrent users. Covers common system designs (TinyURL, feeds, chat) and back-of-the-envelope estimation. For data fundamentals, see ddia-systems. For resilience, see release-it.
Trace the origins of theories and reinterpret cases. Review experiential judgments in presentations, courses, copy, or methodologies: first determine whether the proposition holds, then identify and verify credible theories, reinterpret the case through those theories, and state their limits. Use to find support for an original theory, add theoretical sources to a presentation, assess whether a statement can be elevated to a theory, or check for forced attribution to a famous person. Focuses on a single theoretical anchoring and does not default to broader historical analogy research.
Interactively discovers requirements and designs holistic, multi-product system architectures, solution blueprints, and deployment recommendations for complex workloads on Google Cloud. Use when designing end-to-end cloud solutions, selecting and integrating Google Cloud services, generating architecture diagrams, or conducting requirements discovery for new cloud workloads or migrations. Don't use for single-product tasks (use product-specific skills), initial onboarding or authentication (use google-cloud-recipe-*), Well-Architected Framework reviews or audits (use google-cloud-waf-*), or workloads covered by specialized solution skills.
AI image generation for SEO assets: OG/social preview images, blog hero images, schema images, product photography, infographics. Powered by Gemini via nanobanana-mcp. Requires the banana extension installed. Use when user says "generate image", "OG image", "social preview", "hero image", "blog image", "product photo", "infographic", "seo image", "create visual", "image-gen", "favicon", "schema image", "pinterest pin", "generate visual", "banner", or "thumbnail".
Turn a token name into a vetted, sized buy order. Two things here are exclusive to this skill: resolving a NAME or symbol to the one right contract among its copycats, and sizing an order — amount, slippage, gas, position. No sibling does either. USE THIS SKILL WHEN a buy is being prepared, which shows up as a NAME or symbol, or an AMOUNT, or both — including the plainest possible ask with no mention of checking: 我想买 200u 的 PENGU, 帮我买 500 刀的 BONK, 买 1 个 SOL 的 WIF, 帮我买点 dogwifhat, 想梭 100u 的 XX, PENGU 现在能买吗, PENGU 能不能买, 能不能冲, 值不值得进, 确认是正主不是仿盘再买, buy me $500 of PENGU — and including a bare contract address once an amount arrives with it. DO NOT USE THIS SKILL for a bare contract address with no name to resolve and no amount to size (「这个地址能不能买」, 打个分, 尽调, 有没有貔貅, rug check, is this token safe): that one belongs to gmgn-contract-dd, which is the skill that returns the 0-100 safety verdict. This skill never computes a second verdict of its own — it CALLS gmgn-contract-dd and defers to it — so there is nothing to gain by taking that ask, and the user gets an order card they never asked for. Come back here the moment they name an amount. A buy question about a LAUNCHER rather than about a token — 「这个 dev 的新盘能不能买」, 「他下一个盘值不值得冲」, "should I buy his next launch", "will this dev rug at open" — is gmgn-dev-score: it scores the creator's own record, and there is no token name to resolve or amount to size yet. Come back here once they name the coin. Naming a token with no buy intent at all is gmgn-market search. gmgn-swap is where this skill ENDS, not a rival for it: gmgn-swap signs and submits, and this skill never touches a private key and never places an order, so a plain buy request starts HERE and reaches gmgn-swap only after the user confirms the order card. Note that gmgn-swap cannot start from a name either — its --output-token is a contract address and the only names it resolves are the currencies SOL/BNB/ETH/USDC — so 帮我买点 dogwifhat has to come here regardless of who is asked. Go straight to gmgn-swap only when the user says the pre-buy check is unwanted (skip the checks, 直接买, 不用尽调, 我很急), or for what this skill does not do at all: selling, percentage sells, limit orders, stop loss, take profit, trailing orders, multi-wallet batch trading, order status, gas-price lookups. The split is by what the ask still needs done, not by who executes: a name needs resolving and an amount needs sizing, so both start here; a bare address needs only a verdict, so it is gmgn-contract-dd's; a signed transaction is always gmgn-swap's.
Use this skill when loading and managing resources in PixiJS v8. Covers Assets.init, Assets.load/add/unload, bundles, manifests, background loading, onProgress, caching, spritesheets, video textures, web fonts, bitmap fonts, animated GIFs, compressed textures, SVG as texture or Graphics, resolution detection, per-asset data options, and forcing a specific loader with the parser field (for extension-less URLs). Triggers on: Assets, Assets.load, Assets.init, loadBundle, manifest, backgroundLoad, Spritesheet, Cache, LoadOptions, unload, parser, loadParser, loadWebFont, loadBitmapFont, loadVideoTextures, GifSource, VideoSourceOptions.
OKX-aggregated DeFi discovery and execution — for users who want OKX to find and route to the best protocol WITHOUT naming a specific DApp. **If the user names ANY third-party protocol/DApp as destination (Aave, Lido, PancakeSwap, Uniswap, Curve, Compound, Morpho, Pendle, Kamino, Raydium, Hyperliquid, Polymarket, etc.), route to okx-dapp-discovery — NOT here, even if the verb (deposit, stake, add liquidity, borrow, claim) matches.** DApp-agnostic triggers: 'invest in DeFi', 'earn yield on stablecoins', 'find best APY', 'deposit/stake for yield', 'search DeFi products', 'redeem/withdraw position', 'claim DeFi rewards', 'borrow against asset', 'repay loan', 'add/remove CLMM liquidity', 'APY/TVL history', 'CLMM depth chart'. Also covers DeFi investing, yield farming, lending, borrowing, staking, liquidity pools, APY/TVL trends. Do NOT use for: DEX swaps (okx-agentic-wallet), token prices (okx-dex-market), wallet balances (okx-agentic-wallet), positions-only views (okx-defi-portfolio).
Provides instructions to implement, integrate, or configure Google Mobile Ads (GMA) banner ads in Android, iOS, or Unity mobile applications. Use when the task involves setting up banner ads in a mobile application. Don't use for other ad formats like interstitial or rewarded ads.
Provides instructions for implementing, integrating, or configuring Google Mobile Ads (GMA) SDK interstitial ads in Android, iOS, or Unity mobile applications. Use when the task involves setting up interstitial ads. Don't use for "rewarded interstitial" ads.
Instrument any app with OpenTelemetry and ship metrics / logs / traces to Grafana Cloud or self-hosted Mimir / Loki / Tempo / Pyroscope. Covers SDK auto-instrumentation for Go, Java (Grafana JVM agent), Python (`opentelemetry-instrument`), Node.js, .NET (`Grafana.OpenTelemetry`), Beyla eBPF for zero-code; Grafana Cloud OTLP gateway + Basic-auth (instanceID + API key, base64); env-var config (`OTEL_EXPORTER_OTLP_*`, `OTEL_RESOURCE_ATTRIBUTES`); Alloy / OTel-Collector pipelines; Kubernetes Operator inject-annotations; and head + tail sampling. Use when instrumenting a service, pointing OTLP at Grafana Cloud, switching from Jaeger / Datadog / New Relic, choosing head- vs tail-sampling, or debugging "spans aren't showing in Explore" — even when the user says "auto-instrument my Java app", "send traces to Grafana", "what env vars do I set", "OTLP endpoint", or "Operator inject" without naming OpenTelemetry.
Provides instructions for implementing, integrating, or configuring Google Mobile Ads (GMA) SDK rewarded ads in Android, iOS, or Unity mobile applications. Use when the task involves setting up rewarded ads. Don't use for "rewarded interstitial" ads.
Performs code upgrades, migrations, and transformations using the AWS Transform (ATX) CLI. Use when upgrading language versions, migrating AWS SDKs, migrating frameworks (Angular, Vue.js, Spring Boot, React), upgrading libraries, optimizing performance, migrating x86 to Graviton, analyzing codebases / generating documentation, or defining custom transformations with natural language. Runs locally on a few repositories or at scale across hundreds via AWS Batch/Fargate.
Periodic pass that keeps a project's verification skill and feature map honest: parallel source readers per feature, one live session driving every feature, at most one PR of proven corrections. Use for /maintain-verification-skill or "audit the verify skill".
Gets AWS credentials for CLI/SDK access via `aws login`. Activates when a developer needs to authenticate to AWS for local development, when an AWS operation fails due to missing or expired credentials, or when someone asks about setting up AWS access. Triggers: "set up AWS", "configure AWS", "aws login", "get credentials", "authenticate", "session expired", "token expired", "no credentials", "AccessDeniedException" with no configured credentials.
Relationship analyst combining structural diagnostics (5-layer framework) with psychoanalytic depth (transference, unconscious patterns, resistance). Guides users through dialogue to "see" the real structure of their relationship issues. Use when user says "关系分析", "分析关系", "relationship", "人际关系", or describes a specific relationship problem they want to understand.
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.