Skills
97,905 packaged capabilities and playbooks for agents. Install any of them with one command.
Development
Distributed training orchestration across clusters. Scales PyTorch/TensorFlow/HuggingFace from laptop to 1000s of nodes. Built-in hyperparameter tuning with Ray Tune, fault tolerance, elastic scaling. Use when training massive models across multiple machines or running distributed hyperparameter sweeps.
Production-grade README.md patterns for any project type. Use when creating project documentation, writing README files, or improving existing docs. Covers hero sections, quick start examples, comparison tables, troubleshooting guides, and limitation transparency. Triggers on README, documentation, project setup, open source.
Review software architecture for coupling, cohesion, SOLID principles, API design, scalability, and technical debt. Covers system-level evaluation, architecture decision record review, and improvement recommendations. Use when evaluating a proposed architecture before implementation, assessing an existing system for scalability or security, reviewing ADRs, performing a technical debt assessment, or evaluating readiness for significant scale-up.
Use the RunAPI CLI for one-off artifacts and results, or RunAPI's standard x402 REST flow for wallet-funded Task calls without an API key. Use when an agent needs to discover models or prices, create a paid Task, poll its result, or run a supported CLI service. Use an SDK for app or production integration.
Covers the RWKV (Receptance Weighted Key Value) architecture, an RNN/Transformer hybrid with O(n) inference and no KV cache, including RWKV-7, its parallel GPT-mode training and sequential RNN-mode inference, state passing, fine-tuning with DeepSpeed, and CUDA kernel setup. Use when generating text token by token with constant memory, processing very long contexts of 100K+ tokens, fine-tuning an RWKV model, comparing RWKV memory and speed against Transformers, or debugging RWKV state handling, loading, or out-of-memory errors. Prefer a standard Transformer when peak accuracy matters more than memory, and Mamba for state-space models.
Comprehensive software architecture skill for designing scalable, maintainable systems using ReactJS, NextJS, NodeJS, Express, React Native, Swift, Kotlin, Flutter, Postgres, GraphQL, Go, Python. Includes architecture diagram generation, system design patterns, tech stack decision frameworks, and dependency analysis. Use when designing system architecture, making technical decisions, creating architecture diagrams, evaluating trade-offs, or defining integration patterns
Generates sentence, text, and image embeddings locally with the Python sentence-transformers (SBERT) library, using pre-trained Hugging Face models such as all-MiniLM-L6-v2, all-mpnet-base-v2, and multilingual variants. Covers encoding, cosine similarity, semantic search, batch encoding, fine-tuning, and LangChain/LlamaIndex integration. Use when building embeddings for RAG, running semantic search or similarity scoring, clustering or classifying text, embedding multilingual text without an API, or fine-tuning an embedding model on domain data. For API-based or managed embeddings, use OpenAI or Cohere Embed instead.
Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k sentences/sec), lightweight (6MB memory), deterministic vocabulary. Used by T5, ALBERT, XLNet, mBART. Train on raw text without pre-tokenization. Use when you need multilingual support, CJK languages, or reproducible tokenization.
Guided entry point for using Sentry through your agent. Orients you to your current setup and, for a new project, sets up Sentry end to end with sane defaults — provision a project, install the SDK (errors, tracing, and whatever it enables by default), and confirm real telemetry reaches Sentry. Routes other intents (adding more signals, fixing issues) to the right skill.
Comprehensive skill for Sentry error monitoring and performance tracking. Use when Claude needs to (1) Configure Sentry SDKs for error tracking and performance monitoring, (2) Manage releases, source maps, and debug symbols via CLI, (3) Query issues, events, and metrics via API, (4) Set up alerting and notification rules, (5) Configure sampling strategies and quota management, (6) Deploy self-hosted Sentry instances, (7) Integrate with OpenTelemetry for distributed tracing, or any other Sentry automation task.
Manages shadcn components and projects — adding, searching, fixing, debugging, styling, and composing UI, including chat interfaces. Provides project context, component docs, and usage examples. Applies when working with shadcn/ui, component registries, presets, --preset codes, or any project with a components.json file. Also triggers for "shadcn init", "create an app with --preset", "switch to --preset", and for building a design system or applying a DESIGN.md or brand spec to shadcn/ui.
Take a PR from review to merged — run the repo's code-review skill on it in a loop (review, fix, re-review) until nothing is left to fix, applying ALL fixes on the PR branch, wait for CI green, then squash-merge. Use when asked to "ship it", "ship PR #N", or "review, fix and merge" a PR.
Use Shopware CLI for Shopware project, extension, and account workflows — create and install new projects (project create, project dev install), validate projects or extensions (with --only/--exclude and --format json, junit, github, gitlab for CI), develop, build, upgrade, and troubleshoot. Also use when contributing to shopware/shopware-cli and reasoning about the CLI's user-facing behavior.
Provide implementation guidance for Sileo (npm package "sileo") toast notifications in React apps. Use when adding the Toaster viewport, firing global toasts via sileo.success/error/warning/info/show/action, handling async flows with sileo.promise, dismissing/clearing toasts, configuring positions/offset, theming with light/dark/system modes, and customizing styling via fill, styles overrides, autopilot, roundness, CSS variables, and data-attribute selectors. Also use this skill for swipe-to-dismiss behavior, custom icons, JSX descriptions, reduced-motion support, TypeScript type exports, and understanding Sileo's internal SVG gooey morphing architecture. Make sure to use this skill whenever the user mentions sileo, toast notifications in React, gooey toasts, physics-based notifications, or wants to add any kind of toast/notification system using sileo, even if they don't explicitly name the package.
Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.
Use when migrating from SpacetimeDB 1.0 to 2.0 or fixing code affected by 2.0 breaking changes. Covers reducer callbacks, event tables, event type changes, subscription API, table accessor naming, database-name connections, `sender()` method, update methods, scheduled functions, private codegen, light mode, `CallReducerFlags`, and confirmed reads. Triggers on: upgrade, migrate, 1.0, 2.0, breaking changes.
Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models.
Drive an evidence-driven, iterative product+engineering spec process that produces a full PRD + technical spec (often as SPEC.md). Use when scoping a feature or product surface area end-to-end; defining requirements; researching external/internal prior art; mapping current system behavior; comparing design options; making 1-way-door decisions; negotiating scope; and maintaining a live Decision Log + Open Questions backlog. Triggers: spec, PRD, proposal, technical spec, RFC, scope this, design doc, end-to-end requirements, scope plan, tradeoffs, open questions.
Engage the User to resolve important open decisions in Spec Loop work, including tasks, plans, design updates, work breakdowns, implementation checkpoints, and ADR drafting. Use when the user asks to clarify, compare options, discuss criteria, or stress-test, or when different answers could materially change scope, behavior, policy, constraints, route, Design, Test specification, or ADR content. Provides mandatory decision screening for workflows that require it before dependent Spec Loop content is written or revised. Also use for general grilling when explicitly selected.
Create specs, describe an existing project, update agreed contracts as behavior changes, or review specs against implementation. Infer spec-init, spec-onboard, spec-update, or spec-review from intent and evidence. Use spec-update for the canonical part of an authorized implementation change; keep code and user documentation in their own steps. Activate yourself after completing an authorized change that alters material behavior, compatibility, or a security boundary when specs/ exists; this trigger is owned by the skill and needs no project instructions. Russian discovery terms: спецификация проекта, обновить спецификацию, проверить спецификацию.
Pure-web protocol reverse skill: turn hostile browser clients into browser-free Python collectors. Auto-judge intake and tool path: artifact-only when samples suffice; fresh web live-target needs sequential fingerprint-baseline then debugger-trace (default chrome-devtools then js-reverse; if DP is already required js-reverse may own both roles on one host and must not launch a second Chrome; replay identity locks to installed curl impersonate; unknown risk-control does not bulk-add unproven extra routes; Camoufox/managed baseline only on fingerprint pressure; silent-value-capture is a first-class debugger-trace means when attach is missing or page-world capture is insufficient/observer-toxic). Continuation reuses the current gate. Route only mounted web MCP families. Out of scope as primary path: APK, native app, mini-program. Use for hostile web sign, token, cookie, WebSocket, GraphQL, protobuf, response-decode, fingerprint, WASM, challenge-bootstrap, dynamic-font, or protocol-collector flows
Comprehensive Swift development for building, testing, and deploying iOS/macOS applications. Use when Claude needs to: (1) Build Swift packages or Xcode projects from command line, (2) Run tests with XCTest or Swift Testing framework, (3) Manage iOS simulators with simctl, (4) Handle code signing, provisioning profiles, and app distribution, (5) Format or lint Swift code with SwiftFormat/SwiftLint, (6) Work with Swift Package Manager (SPM), (7) Implement Swift 6 concurrency patterns (async/await, actors, Sendable), (8) Create SwiftUI views with MVVM architecture, (9) Set up Core Data or SwiftData persistence, or any other Swift/iOS/macOS development tasks.
Contributing to the Twenty codebase itself (twentyhq/twenty server internals), not for building apps on top of Twenty. Define types, entities, and central constant registrations for syncable entities in Twenty's workspace migration system. Use when creating new syncable entities, defining TypeORM entities, flat entity types, or registering in central constants (ALL_ENTITY_PROPERTIES_CONFIGURATION_BY_METADATA_NAME, ALL_ONE_TO_MANY_METADATA_RELATIONS, ALL_MANY_TO_ONE_METADATA_FOREIGN_KEY, ALL_MANY_TO_ONE_METADATA_RELATIONS).
Create and configure ASP.NET Core Stepper controls to support multi-step workflows and progressive user interfaces. This skill should be applied when implementing step navigation, wizards, progress trackers, form flows, checkout processes, or any sequential step-based UI. It is also relevant whenever requirements involve steppers, step validation, step navigation, linear flows, or step-based state management.
Implements Syncfusion Flutter Barcode Generator (SfBarcodeGenerator) for generating 1D and 2D barcodes in Flutter apps. Use when working with QR codes, Data Matrix, Code128, EAN, UPC, or other machine-readable barcode formats. This skill covers barcode types, customization, sizing, text display, and integration into product labels, ticketing, or inventory systems.
Implements Syncfusion Flutter Funnel Chart (SfFunnelChart) for proportional and stage-based data visualization in Flutter apps. Use when working with conversion funnels, sales pipelines, or process-stage visualizations. This skill covers series configuration, segment exploding, gap ratio, data labels, legends, tooltips, and customization.
Implements Syncfusion Flutter Gauge widgets (SfLinearGauge, SfRadialGauge) for data visualization and measurement displays in Flutter apps. Use when building speedometers, progress indicators, KPI dashboards, or radial/linear measurement UIs. This skill covers gauge axes, pointers, ranges, annotations, and customization for both linear and radial gauge types.
Build carousel components for displaying image galleries, news headlines, and rotating content. Use this skill when implementing image sliders, automatic content rotation, or carousel-based galleries. Covers navigation controls, slide transitions, animation effects, and indicators.
Guides implementation of the Syncfusion TypeScript Markdown Converter (@syncfusion/ej2-markdown-converter). Use this skill when the user needs to convert Markdown text to HTML, use the toHtml method, configure MarkdownConverterOptions (async, gfm, lineBreak, silence), or integrate the Markdown Converter with the Syncfusion Rich Text Editor for live preview or side-by-side editing. Trigger when user mentions markdown to HTML conversion, MarkdownConverter, toHtml, ej2-markdown-converter, markdown preview, or RTE markdown mode.
Implements the Syncfusion Javascript (ES6) is case-sensitive PDF Viewer (PdfViewer) for embedding, configuring, and loading PDF documents. Use this when rendering PDFs in a TypeScript application using the EJ2 class-based approach, embedding viewer controls, or generating TypeScript and HTML code for PDF display and interaction.
Create and configure Syncfusion TypeScript Timeline control for displaying sequential events and chronological content. Use this skill when implementing timeline layouts, milestone tracking, process flows, progress indicators, or visualizing ordered sequences. This skill covers project milestones, career progression, order status tracking, itinerary planning, and procedural workflow visualization using Syncfusion Timeline.
Use this skill for UWP Application needing Excel-like UI using the Syncfusion Spreadsheet Component. Trigger for creating, viewing, editing Excel (.xlsx, .xls) files; using formulas, filtering, sorting, or cell formatting. Also trigger when users reference spreadsheet files ("open xlsx", "load Excel file", "add Syncfusion spreadsheet"). Do NOT trigger for standalone file processing without UI components.
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.