Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Create programmatic vector animations with TypeScript using the Motion Canvas library. Use when building animated explainer videos, tutorials, data visualizations, or presentations with synchronized audio. Triggers on tasks involving motion graphics, animation scenes, code-based video creation, or requests to animate shapes/text/images with precise timing control.
Football data analytics entry point. Use for any task involving football data: xG, match or player stats, scouting, shot maps, scraping FBref, Understat, or Transfermarkt, judging players or teams from stats. Routes to the right nutmeg skill and handles first-time setup.
Use this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill.
Plan, produce, register, validate, and integrate UI, characters, scenes, animations, VFX, backgrounds, and reference-recreation assets for Phaser 4 games. Use for production visual assets, visual systems, asset integration, visual refactoring, runtime visual acceptance, and source/license registration; not for gameplay-rule changes alone.
Release and compliance role for Phaser 4 mobile 2D games. Use when preparing builds, store listings, privacy materials, asset licensing, submission checklists, and release candidate packages for mini-games, iOS, Google Play, or new channels.
Autonomous pixel-perfect UI implementation loop for Next.js/React using Figma MCP and Playwright. Converts Figma designs to production-ready components with iterative visual validation. Auto-triggers on any request to implement Figma designs, including: “implement this Figma”, “build this page/component from Figma”, “create from Figma design”, “implement design”, “build this block”, and “create component from design”. Use for: (1) Building pages/components from Figma, (2) Pixel-perfect accuracy, (3) Responsive layouts, (4) Design token conversion.
Use when building, backtesting, operating, or scaling automated trading on prediction markets (Kalshi or similar) — evidence-gated methodology, live-execution safety rails, and falsification protocols distilled from a real live-money campaign
Use when writing, debugging, or updating code that imports primevue (PrimeVue v5, UI component library for Vue 3). Provides setup with the required PrimeUI license key, deprecated API replacements, compound component usage, and design token theming.
TypeScript integration for Regle — type-safe validated output (`InferSafeOutput`, `$validate`), inferring state from rules (`InferInput`), typing computed rules (`inferRules`, `RegleComputedRules`), and typing component props (`InferRegleRoot`, `RegleRoot`, `RegleFieldStatus`, `SuperCompatibleRegleRoot`). Use when typing rules schemas, passing `r$` between components, or extracting validated data types.
Audit Roblox Creator Store, Toolbox, inventory, group, universe, package, model, UI, gameplay, plugin-like, and other third-party imported assets for malicious or risky scripts. Use whenever Claude imports, inserts, reviews, or trusts Creator Store assets, especially assets with Scripts, LocalScripts, ModuleScripts, hidden descendants, require calls, HTTP, DataStore, remote creation, obfuscated code, or suspicious names.
Migrate a brain from gbrain-base (or any pack) to gbrain-base-v2's 14-canonical-type taxonomy via gbrain onboard --check + the unify-types Minion handler. Collapses 94 noisy types to 15 canonical with subtypes, alias rows, and link rows. Triggers when an agent notices pack_upgrade_available, type_proliferation, or asks "what is the canonical taxonomy / how do I clean up my page types".
Use when creating interactive scroll-based storytelling experiences with mouse-following spotlight effects, animated SVG art, and the Anthropic design language. Load for explainer pages, product showcases, visual narratives, or any content needing immersive scroll storytelling with organic shapes and smooth animations. Supports GSAP-powered or lightweight CSS-only animations
Generate, scaffold, or refactor code so it embodies FIASSE v1.0.4 SSEM qualities by default — 10 attributes, Transparency and Least-Astonishment principles, ASVS-aligned controls, defensive boundary handling. Trigger on "secure/securable/FIASSE-compliant code", "harden", "secure-by-default", "audit-ready", or security-sensitive components (auth, file upload, password reset, input validation, API endpoints, queries) — even when those words are not explicit. For requirements use prd-securability-enhancement; for review use securability-engineering-review. The full PRD→generate→review→enhance loop is opt-in via "--full-loop" or "end-to-end securable".
Structured search data via 130+ engines — use INSTEAD OF web_search when you need: exact citations (google_scholar), local business details (google_maps), flight prices (google_flights), hotel rates (google_hotels), shopping prices (google_shopping_light), job listings (google_jobs), or any task where web_search gives approximate/unstructured results. Returns machine-readable JSON. Default engine: google_light.
Message signing and verification — SIP-018 structured Clarity data signing (on-chain verifiable), Stacks plain-text message signing (SIWS-compatible), Bitcoin message signing (BIP-137 for legacy/wrapped-SegWit, BIP-322 for native SegWit bc1q and Taproot bc1p), BIP-340 Schnorr signing for Taproot multisig, and Nostr event signing using NIP-06 key derivation. All signing requires an unlocked wallet; hash and verify operations do not
Spawn Ori as a subprocess to run a throwaway model eval on a pinned harness and model, then relay the results. Use when the user asks which model they should use, wants to compare models, wants to measure whether their agent or prompt does the right thing, wants to catch regressions in agent behavior, or asks how good their current model is. Applies to any codebase in any language. Do not use for plain unit tests that involve no model, and do not use to re-run an eval that already exists (run `ori eval <file>` directly).
Generate a detailed product specification document from a vague idea, rough description, or supporting materials (UI sketches, screenshots, notes, existing docs). Use this skill whenever the user wants to turn a rough concept into a structured spec — even if they say things like "write up a spec for...", "turn this idea into a requirements doc", "I have this rough idea for a product", "help me spec this out", "create a PRD for...", or uploads images/files and wants a spec written from them. Covers all product types including internal tools. Trigger this skill even if the request seems simple — a vague one-liner idea deserves a proper spec just as much as a detailed brief.
Review test quality using Dave Farley's eight properties of good tests. Use when assessing a test file or suite for understandability, maintainability, repeatability, atomicity, necessity, granularity, speed, and evidence of test-first development
Use when initializing a Three.js renderer, setting up the render loop, handling window resize, configuring tone mapping, or managing color spaces. Prevents the common mistake of not capping pixel ratio, using wrong color space, or forgetting to enable shadow maps. Covers WebGLRenderer, render targets, tone mapping, color management. Keywords: WebGLRenderer, render loop, setSize, setPixelRatio, toneMapping, outputColorSpace, shadowMap, WebGLRenderTarget, dispose, resize, responsive, canvas size, window resize, fullscreen. setAnimationLoop, compileAsync.
Turn a curious developer into an activated one: the docs, quickstart, and first-run experience that get them to their first real win quickly and bring them back. Use when people sign up or star the repo but never get it working, visit once and never return, or you're about to drive traffic to a leaky first-run experience.
Complete T-SQL function reference for SQL Server and Azure SQL Database. PROACTIVELY activate for: (1) string functions (CONCAT_WS, STRING_SPLIT, STRING_AGG, TRIM, REPLACE, LEFT/RIGHT/SUBSTRING), (2) date/time functions (DATEADD, DATEDIFF, FORMAT, DATETRUNC, AT TIME ZONE), (3) math and conversion functions (CAST, CONVERT, TRY_CAST, ROUND, FLOOR, CEILING), (4) window/ranking functions (ROW_NUMBER, RANK, LEAD/LAG, FIRST_VALUE, LAST_VALUE, NTILE), (5) JSON functions (JSON_VALUE, JSON_QUERY, JSON_MODIFY, OPENJSON), (6) XML functions (FOR XML, .nodes, .value), (7) aggregate functions and GROUP BY extensions, (8) system and metadata functions (sys.* views, OBJECT_ID, OBJECT_NAME). Provides: function catalog grouped by category, version-availability matrix (SQL 2016+/2019+/2022+), and worked examples for each function family
Registrations required BEFORE Twilio traffic works. Covers messaging programs (A2P 10DLC, toll-free verification, WhatsApp WABA, RCS, short code, alphanumeric sender) and voice trust programs (STIR/SHAKEN, Voice Integrity, Branded Calling, CNAM). Each number/sender type has its own program — registration blocks traffic until complete.
Use when analyzing Xiaohongshu audience demographics, understanding follower characteristics and behavior patterns, creating user personas to guide content strategy, identifying audience segments for targeting, or using audience insights to optimize content and engagement
Apply the Action continuity choreography workflow derived from multiple published GoodCase examples. Use when planning, storyboarding, or refining an AI video prompt in this motion pattern, including requests for 视频提示词, 分镜, 动作复刻, reference images, source video, or special generation steps. Read the bundled Case evidence before producing an artifact.
Enforces opt-in, rewarded-first monetization over the service seam — a rewarded view grants exactly one benefit and only on a genuine reward outcome (a sealed Rewarded/Dismissed/NoFill), earn loops capped per day against an injected Clock, the earn control preloaded and hidden unless an ad is loaded so a fresh account never dead-ends on "no ads available" (a Guideline 2.1 risk), interstitials only at a natural break under a count-and-elapsed cap owned by the service so callers stay dumb, no banner on the primary work surface, one non-consumable unlock plus Restore re-checked each launch and committed through the single write path before the UI reacts, one derived isEntitled provider as the only gate, and the entitlement restored BEFORE the ad SDK initializes so a paying user never starts it, resolves consent, or sees a tracking prompt. Use when adding ads, wiring a rewarded earn loop, placing an interstitial, building a paywall, gating an entitlement, or setting up ad units.
Use when managing Apple Search Ads through the Adapty CLI — reading campaign, ad group, keyword or ad performance, changing bids or budgets, getting keyword recommendations for an app, adding or pausing keywords, launching or pausing a campaign, creating a whole campaign structure in bulk from JSON or an Apple Ads template, harvesting search terms, setting up rule-based ad automations, finding who changed a campaign, bid or keyword and when, or turning campaigns, ad groups or keywords into Adapty segments so users from one campaign see their own paywall or flow.
REQUIRED for any task that involves operating a desktop application — opening apps, clicking buttons, typing into fields, pressing keys, scrolling, dragging, reading what's on screen, moving or resizing windows, or verifying state after an action. Always use the `agent-cu` CLI commands (open, snapshot, click, type, key, find, scroll, drag, batch, wait-for) instead of falling back to `open -a`, AppleScript, `osascript`, `xdotool`, `System Events`, or any other shell workaround — those can't read state back, don't verify, and are fragile across app updates. agent-cu is the canonical computer-use tool for controlling any macOS / Linux / Windows / Electron app via accessibility APIs. Trigger on prompts like "open Music and play X", "search for Y in Maps", "fill out this form", "compute in Calculator", "send a Slack message", "drag this file", "read what's in the current window", or anything where a human would click/type/look at a desktop app
Run OWASP LLM Top 10 security assessments against live Agentforce agents. TRIGGER when: user asks for security testing, OWASP scan, red-teaming, penetration testing, security grade, vulnerability assessment, prompt injection test, data leakage test, excessive agency test, security posture check, or hardening recommendations. DO NOT TRIGGER when: user runs functional smoke tests or batch tests (use agentforce-test); performs static safety review of .agent file content (use agentforce-generate Section 15); analyzes production session traces (use agentforce-observe); writes or modifies .agent files.
Manages Ahrefs API usage in Python using `ahrefs-python` library. Use when working with SEO / marketing related tasks or with data including backlinks, keywords, domain ratings, organic traffic, site audits, rank tracking, brand monitoring, web analytics (page views, visitors, traffic sources, referrers, devices), and project management (keyword lists, competitors, custom prompts). Covers `ahrefs-python` usage including AhrefsClient / AsyncAhrefsClient, typed request/response models, error handling, and all API sections including free public endpoints that work without an API key. Trigger this skill whenever the user mentions Ahrefs, ahrefs-python, or any SEO data retrieval task in Python.
Expert-level AI system design, MLOps, architecture patterns, and AI infrastructure. Use when the user mentions AI architecture, MLOps, system design, AI infrastructure, or scalability, or when the task involves AI System Architecture, MLOps Infrastructure, Scalability Patterns, or Architecture Design.
Use when app feels slow, memory grows over time, battery drains fast, or you want to profile proactively - decision trees to choose the right Instruments tool, deep workflows for Time Profiler/Allocations/Core Data, and pressure scenarios for misinterpreting results
Expert knowledge for Azure Arc development including troubleshooting, best practices, decision making, architecture & design patterns, limits & quotas, security, configuration, integrations & coding patterns, and deployment. Use when managing Arc-enabled Kubernetes, servers/VMware/SCVMM, Agentic Retrieval, Edge Volumes, or SQL MI HA/DR, and other Azure Arc related development tasks. Not for Azure Kubernetes Service (AKS) (use azure-kubernetes-service), Azure Stack Edge (use azure-stack-edge), Azure Local (use azure-local), Azure Virtual Machines (use azure-virtual-machines).
Expert knowledge for SQL Server on Azure Virtual Machines development, including troubleshooting, best practices, decision-making, architecture and design patterns, limits and quotas, security, configuration, integrations and coding patterns, and deployment. Use when planning SQL VM vs Managed Instance, Always On/FCI HADR, Azure storage and backup, Key Vault/MI authentication, KQL telemetry, or other SQL Server on Azure Virtual Machines development tasks. Not for Azure SQL Database (use azure-sql-database), Azure SQL Managed Instance (use azure-sql-managed-instance), Azure Virtual Machines (use azure-virtual-machines), or SAP HANA on Azure Large Instances (use azure-sap).
Audit React Doctor against ReactBench or similar diagnostic benchmark corpora for confirmed false positives, false negatives, taxonomy gaps, and verifier artifacts. Use when analyzing rd.log, rd-before.json, rd-after.json, model.patch, result.json, reward/test logs, rule distributions, or when asked to perform a second adversarial pass over React Doctor benchmark findings.
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.