Skills
98,639 packaged capabilities and playbooks for agents. Install any of them with one command.
Development
Use this skill to gate Python code (floors 3.10+, rules verified through 3.14) with a pass/fail adversarial review — a single blind reviewer subagent judges a diff or file set against 20 decidable rules hunting two failure modes. First, code modern Python makes unnecessary — branch ladders over match/registries, hand-written init/repr/eq over dataclasses, TypeVar ritual over PEP 695, typing.Optional over PEP 604 unions, os.path over pathlib, hand-rolled stdlib batteries, deprecated utcnow, orphan create_task over TaskGroup. Second, legacy-pattern propagation — single-implementation ABCs, pass-through layers, single-method classes, concrete-inheritance reuse, boolean-forked functions, shapeless payloads and parameter clumps — judged as if greenfield; consistency with legacy code is not PASS evidence. A version probe reads the target's Python floor, marks rules above it N/A, and fetches the official what's-new delta when the floor exceeds the verified version. Verdicts only, never fixes.
Use this skill to gate Elixir code built on the Reactor orchestration library (~> 1.0) with a pass/fail adversarial review — a single blind reviewer subagent judges a diff or file set against 28 decidable rules covering saga compensation and undo (side effects without undo, cleanup in the wrong callback, compensate returns that roll back vs continue, non-idempotent undo), retry discipline (uncapped retries under the max_retries infinity default, retrying business failures, missing backoff), dependency and data flow (lexical-order assumptions, context smuggling, missing return), step contracts (invalid run/3 returns, halt misused as failure, guard/where confusion, side effects in inline fns), composition (Reactor.run inside steps instead of compose, Enum loops over map steps, case over switch, unbounded recurse), concurrency (serial-by-default map, sandbox tests left async, process-context loss), and middleware contracts. Verdicts only, never fixes.
For any request to run, execute or evaluate code (such as run this code, execute script, draw a chart in Python, run code/draw a chart/execute a script). Creates a code_latest sandbox via the AgentBay SDK, calls run_code to execute and returns the result. Supports Python, JavaScript, R and Java.
Apple data persistence, data flow, networking, and utility reference. SwiftData / Core Data / Foundation / Combine. @Model, ModelContainer, ModelContext, @Query, Predicate, FetchDescriptor, SectionedResults, ResultsObserver, NSManagedObject, NSFetchRequest, NSFetchedResultsController, NSPersistentContainer, NSPersistentCloudKitContainer, NSStagedMigrationManager, URLSession, JSONDecoder, JSONEncoder, Publisher, sink, Future, AsyncPublisher, @Published, AnyCancellable
App Store Connect API 4.5.1 (REST) app distribution reference. JWT authentication, API key creation and revocation, Apps, Builds, Bundle IDs, Bundle ID Capabilities, Certificates, Devices, Provisioning Profiles, Users, User Invitations, Merchant IDs, Pass Type IDs, TestFlight (Beta Groups / Testers), Sandbox Testers, App Review, Review Submissions, Customer Reviews, Custom Product Pages, Phased Release, In-App Purchases, Subscriptions, Subscription Groups, App Price Points, Promotional Offers, Win-back Offers, Sales Reports, App Asset Library (screenshots / app previews), Webhooks / Webhook Events / Deliveries, Xcode Cloud (Workflows, Build Runs, Artifacts), Alternative Distribution (EU marketplaces / Web Distribution), Game Center (Achievements, Leaderboards, Challenges), Analytics / diagnostic reports.
Use when creating professional architecture diagrams, cloud infrastructure visuals, network topologies, Kubernetes cluster diagrams, or microservices architecture diagrams as PNG/SVG images using Python Diagrams library with real provider icons (AWS, Azure, GCP, K8s, OnPrem, Generic)
Expert knowledge for Azure Durable Task development, including best practices, decision-making, architecture and design patterns, limits and quotas, configuration, integrations and coding patterns, and deployment. Use when building Durable orchestrations with Azure Storage/SQL/Netherite, versioning deployments, using instance APIs, or handling other Azure Durable Task-related development tasks. Not for Azure Functions (use azure-functions), Azure Logic Apps (use azure-logic-apps), Azure App Service (use azure-app-service), or Azure Service Fabric (use azure-service-fabric).
Use when designing Bevy 0.20 persistence with `DynamicWorldBuilder`, `DynamicWorld::serialize`, `WorldAssetRoot`, `IoTaskPool`, or `SettingsPlugin`: stable IDs, versioned save DTOs/migrations, checksums and atomic writes, snapshots plus sparse deltas, native/browser storage, and safe ECS serialization boundaries.
Use when turning voxel edits into production Bevy 0.20 remesh work with `AsyncComputeTaskPool`, `poll_once`, `Mesh3d`, and Rapier `Collider`: dirty/halo fan-out, revision tokens, stale-result rejection, coalescing, priority, bounded uploads, or atomic swaps.
BlindPay integration for payins/payouts and stablecoin on/off-ramps (ACH, wire, SWIFT, Pix, SPEI, SEPA, USDC/USDT), customer KYC/KYB, bank and virtual accounts, managed or blockchain wallets, cross-chain transfers, payables (boletos, PIX QR codes, invoices), partner fees, or webhooks.
This skill should be used when the user asks about Bootstrap components, "how to create a modal", "navbar not collapsing", "carousel autoplay", "responsive card grid", "toast notifications", "dropdown menu", "accordion FAQ", "offcanvas sidebar", "tab navigation", "tooltip not showing", "popover not working", Bootstrap accordion, alerts, badges, breadcrumb, buttons, button groups, cards, carousel, close button, collapse, dropdowns, list group, modal, navbar, navs and tabs, offcanvas, pagination, placeholders, popovers, progress, scrollspy, spinners, toasts, tooltips, or needs help implementing any Bootstrap UI component.
Use when writing application code in Node.js / JavaScript / TypeScript — or any non-JVM language (Go, PHP, Ruby, Rust, C/C++) — that connects to a data source through a CData driver, after the connection and SQL have been validated with the cdata-cli discovery skill. Covers the CData ODBC Driver edition: confirming the system-installed driver, configuring a DSN or a DSN-less connection string, and connecting via the language's ODBC bindings (e.g. the Node `odbc` package). This is the cross-language build-phase companion to cdata-cli — the fallback for any environment that speaks ODBC.
Provides a workflow for building and reviewing high-quality React and React Native UI with the Coinbase Design System (CDS). Use this skill to create UI or make visual changes to existing UI with CDS components, tokens, style props, icons, and illustrations. Use it for explicit CDS-adherence audits and reviews, not non-visual logic, data, analytics, navigation, tests, generated code, or refactoring. Examples: "build this screen", "fix spacing on this component", "replace this UI with CDS components", "perform a CDS audit on our changes", "check our codebase for CDS adherence", "does this feature use CDS well?"
Provides expertise on Chroma Cloud integration for semantic search and hybrid search applications. Use when the user is working with Chroma Cloud, CloudClient, managed collections, Schema(), Search(), hybrid search, or Chroma Cloud CLI workflows.
Google quantum computing framework. Use when targeting Google Quantum AI hardware, designing noise-aware circuits, or running quantum characterization experiments. Best for Google hardware, noise modeling, and low-level circuit design. For IBM hardware use qiskit; for quantum ML with autodiff use pennylane; for physics simulations use qutip.
Orchestrator drives, Codex codes. Execute an approved plan one task at a time: your agent (Claude Code, etc.) sequences the work, briefs OpenAI Codex to write ALL product code, reviews every diff, runs the test gate BEFORE each commit, commits one task per commit, and opens exactly ONE PR at the end. Use when the user says 'codex-build', 'have codex code it', 'you orchestrate, codex codes', or hands over a plan for step-by-step implementation.
Review a completed code change for behavior-preserving simplification, local correctness, and maintainability. Apply it after a scoped task or when the user asks to simplify a concrete diff. Do not use as a full feature review or to choose the delivery reviewer plan.
Configure session keys and policies for Cartridge Controller to enable gasless, pre-approved transactions. Use when defining contract interaction policies, setting spending limits, configuring signed message policies, or implementing error handling for session-based transactions. Covers SessionPolicies type, policy definitions, verified sessions, and error display modes.
Best practices for using Cursor—rules, commands, skills, subagents, ignore files, Agent security, workflows, and community resources. Use when setting up Cursor, initializing or creating the .cursor folder, writing .cursor/rules or AGENTS.md, creating commands or skills, configuring .cursorignore, working with Agent, discovering rules or MCPs (e.g. cursor.directory), making codebases cursor-compatible, or asking about Cursor workflows, TDD, git commands, or large codebases.
Distributed computing for larger-than-RAM pandas/NumPy workflows. Use when you need to scale existing pandas/NumPy code beyond memory or across clusters. Best for parallel file processing, distributed ML, integration with existing pandas code. For out-of-core analytics on single machine use vaex; for in-memory speed use polars.
Comprehensive codebase reading engine. Systematically reads actual source code line by line through a 6-phase protocol — scoping, structural mapping, execution tracing, deep reading, pattern synthesis, and structured reporting. Source code is the source of truth. Use when needing to truly understand how code works, not just what documentation claims.
Translate a design (Figma mockup, spec, screenshot, or verbal description) into code with pixel- and motion-level fidelity, using the browser's rendering pipeline as a constraint: exact spacing/timing via tokens, GPU-safe animations, no layout thrashing, 60fps interactions. Use when the user says 'implement this design', 'make it match the mockup', 'it doesn't look like the Figma', 'the animation stutters/drops frames', 'set up design tokens', or when a diff between design and implementation needs to be closed. Produces the faithful implementation plus a fidelity report (design value → implemented value → any deviation and why) and performance notes.
Validate that claims in documentation match codebase reality. Extracts verifiable assertions (file paths, commands, function references, behavioral claims, dependencies) from markdown docs and checks them against the actual project. Use after code changes, before releases, or when docs feel untrustworthy.
Create a Technical Specification (SPEC) - Layer 6 of the SDD flow, defining component interfaces, data models, and behavior contracts. Use when ADR decisions are settled and you need an implementation-ready spec before TDD. Single-document authoring primitive; for end-to-end or batch generation the autopilot (`doc-spec-autopilot`) drives this skill.
Periodically audit a repository's docs for staleness against the implementation: run a cheap mechanical scan (dead source-path references, dead links, leftover DEPRECATED/TBD banners, doc-vs-code freshness gap, git-invisible ignored leftovers, completed-but-undeleted plan files), then dispatch parallel read-only agents to verify only the suspicious docs against source, and finally propose update / delete / migrate-to-ADR dispositions for user approval. Use on a monthly or quarterly docs review, after large refactors or renames, or when asked whether docs are stale.
Coordinate Elastic Stack docs releases: classify versions, route 8.x vs 9.x PRs, edit elastic/dev tracking issues, handle same-GA supersession, draft Slack #docs reminders, and watch docs-internal-workflows deploys. Use when the user says "how's my release", "how's my release doing", or asks for release status; also when coordinating docs releases, opening coordinator PRs, editing elastic/dev issues, drafting release reminders, or working with assembler.yml, versions.yml, or conf.yaml.
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.