Skills
98,873 packaged capabilities and playbooks for agents. Install any of them with one command.
Development
Runs the deterministic build_runner codegen pass (drift_dev, freezed, json_serializable, riverpod_generator) as one pinned command with --delete-conflicting-outputs, always before flutter analyze, never hand-editing or force-committing generated *.g.dart / *.freezed.dart / *.drift.dart output, watch only in local dev. Use when regenerating codegen, fixing "missing part file" / "conflicting outputs" / undefined generated-class analyzer errors, after editing a drift table or DAO, a freezed value object, a riverpod Notifier or provider, or a json_serializable model, or after a fresh git clone, branch switch, or pull.
Use when building a SaaS / multi-tenant platform on Spree 6 with the Enterprise `spree_multi_tenant` gem — many isolated merchants on one deployment, each with its own stores, catalog, customers and orders. Covers licensing and install (`spree_enterprise:install`, `spree_multi_tenant:install`), `Spree::Tenant` and row-level `tenant_id` isolation, `SpreeMultiTenant::Base`, `with_tenant` / `without_tenant`, console tenant context, `insert_all`/`upsert_all` pitfalls, background jobs, `Spree.root_domain`, cookie domain and host authorization, global staff vs per-tenant customers. Common phrasings include "multi-tenant Spree", "SaaS on Spree", "each merchant gets a subdomain", "tenant isolation", "records leaking between tenants", "set the tenant in the console", "Blocked hosts lvh.me".
Use when upgrading a Spree 5.x application to Spree 6.0, or when fixing code that broke after that upgrade. Common phrasings include "upgrade to Spree 6", "migrate from 5.6 to 6.0", "Spree 6 breaking changes", "checkout_flow / state machine is gone", "Spree::Adjustment not found", "master variant removed", "PermissionSets NameError", "staff locked out after upgrade", "what replaced Shipment / Taxon / ReturnAuthorization / metafields", "spree_admin gem not found". Covers preconditions (5.6 first, Rails 8.1, Gemfile swaps, roles as data), the ordered 5.6 → 6.0 data backfills, grep recipes for code that must change, and the behavioral review checklist.
End-to-end Stellar development playbook. Covers Soroban smart contracts (Rust SDK), Stellar CLI, JavaScript/Python/Go SDKs for client apps, Stellar RPC (preferred) and Horizon API (legacy), Stellar Assets vs Soroban tokens (SAC bridge), wallet integration (Freighter, Stellar Wallets Kit), smart accounts with passkeys, status-sensitive zero-knowledge proof patterns, testing strategies, security patterns, and common pitfalls. Optimized for payments, asset tokenization, DeFi, privacy-aware applications, and financial applications. Use when building on Stellar, Soroban, or working with XLM, Stellar Assets, trustlines, anchors, SEPs, ZK proofs, or the Stellar RPC/Horizon APIs.
Implement the Syncfusion WPF Maps (SfMap) control for geographical data visualization in Windows Presentation Foundation applications. Use this when displaying geographical data, rendering shape files, or integrating map layers with markers and bubbles. This skill covers ShapeFileLayer, ImageryLayer, map customization, legends, OpenStreetMap/Bing Maps integration, and interactive map features.
Implement Syncfusion WPF SfMarkdownViewer for rendering and displaying Markdown content in WPF applications. Use this when rendering Markdown text, loading Markdown from strings or application-controlled local files, handling hyperlink clicks with URL validation, or displaying Mermaid diagrams with input sanitization. Covers the Source property, HyperlinkClicked events with allowlist-based navigation, and MermaidBlockTemplate customization with input validation. Always validate and sanitize Markdown content before assigning to Source, especially when the content originates from remote or user-supplied sources.
Implements the Syncfusion WPF SfRangeSlider control for selecting numeric ranges with single or dual thumbs. Use when building value-range selection UIs, sliders with ticks/tooltips, or custom orientation and snapping behaviors.
Implements the Syncfusion WPF Smart Text Editor (SfSmartTextEditor), an AI-powered multiline input control with intelligent text completion. Use this when implementing smart text editors, AI-powered text input, or predictive text suggestions in WPF applications. The control provides context-aware typing assistance through integration with AI services including Azure OpenAI, OpenAI, Ollama, Claude, Gemini, Groq, and DeepSeek. Features include dual suggestion display modes (inline/popup), custom phrase libraries for offline suggestions, full text customization, and comprehensive command handling for workflow integration.
Implement and configure the Syncfusion WPF SmithChart (SfSmithChart) component for visualizing impedance and admittance in transmission line applications. Use this when working with high-frequency circuit analysis, RF engineering, or telecommunications data. This skill covers series configuration, dual-axis systems, data markers, legends, appearance customization, and interactive features.
Implement Syncfusion WPF SplitButton (SplitButtonAdv) with dropdown menus, command binding, and data binding. Use this when working with split buttons, dropdown buttons with menus, or button-menu combinations. This skill covers SplitButtonAdv setup, default actions with dropdown menus, command binding, MVVM implementation, and data-bound dropdown items in WPF applications.
Guide for implementing Syncfusion WPF Surface Chart (SfSurfaceChart) for three-dimensional data visualization. Use this when working with 3D surface charts, contour charts, wireframe charts, or mathematical surface plotting with X, Y, Z coordinates. This skill covers surface visualization, topographical charts, and 3D data plotting scenarios for WPF applications.
TanStack Query v5 server state management for React. Covers query/mutation patterns, v4-to-v5 migration (object syntax, gcTime, isPending, keepPreviousData), optimistic updates via useMutationState, SSR/hydration with HydrationBoundary, infinite queries, offline/PWA support, error boundaries with throwOnError, and React 19 Suspense integration. Use when building data fetching, fixing migration errors, debugging hydration mismatches, implementing caching strategies, or configuring mutations.
Record Twilio voice calls correctly. Covers the critical distinction between Record verb (voicemail) and Dial record (call recording), dual-channel for QA, mid-call pause for PCI, Conference recording, and the ConversationRelay workaround. Use this skill whenever you need to capture call audio for compliance, QA, or analytics.
Build multi-channel messaging experiences using Twilio Conversations (classic) API. Covers creating conversations, adding participants (SMS, WhatsApp, chat), sending messages, and handling webhooks. Use this skill to manage persistent multi-party or multi-channel conversations beyond single-message SMS/WhatsApp.
Receive inbound email via SendGrid Inbound Parse webhook. Covers MX record setup, parsed vs raw mode, handling attachments, and common pitfalls. Use when building email-to-app workflows like support ticket creation or email processing pipelines. Requires a SendGrid API key (SG.-prefix) — not applicable to the Twilio Email API (comms.twilio.com).
Use when the user explicitly wants implementation planning before coding, asks to create or revise an implementation plan, supplies requirements with explicit approval evidence, a specification, acceptance criteria, or task list, or has inputs concrete enough to plan but not execute. Do not use for rough unapproved requirements drafting.
Use when a complex or risky code change should be split into Kent Beck's two-step method by first making the change easy through behavior-preserving preparatory refactoring, then making the intended behavior change once the design supports it. This should trigger for requests such as Apply two-step change; Make this risky change safer; Refactor before changing behavior; Separate preparation from behavior change. Part of Plinth Toolkit
Use when designing, implementing, reviewing, testing, or cleaning up feature toggles, feature flags, kill switches, runtime configuration gates, canary controls, staged rollouts, experiments, or temporary compatibility switches in Java enterprise systems. This should trigger for requests such as Design a feature toggle strategy; Review this feature flag; Add a kill switch safely; Test toggle on and off paths; Clean up an expired feature toggle; Plan controlled rollout and rollback behavior. Part of Plinth Toolkit
Use when you need framework-agnostic Hexagonal architecture guidance for Java projects - ports and adapters boundaries, application core independence, driving and driven adapters, dependency direction, infrastructure leakage detection, optional architecture tests, and evidence-backed remediation. This should trigger for requests such as Review Hexagonal architecture; Review ports and adapters boundaries; Enforce dependency direction; Add ArchUnit architecture tests; Detect infrastructure leakage in domain code; Improve ports and adapters boundaries. Part of Plinth Toolkit
Dead code and dependency cleanup, repo-wide: unused deps, unreferenced exports, unreachable code, orphaned files — removed only with tool evidence, in reversible waves. Runs on green main. For diff-scoped cleanup use absolute-simplify. Triggers on "absolute prune", "remove dead code", "find unused deps/exports", "what can we delete", "clean up orphaned files".
Windows and Git Bash compatibility for Azure Pipelines. PROACTIVELY activate for: (1) pipelines that pass on Linux but fail on Windows agents, (2) path conversion issues in pipeline scripts (MSYS, MINGW), (3) shell detection (bash@3, pwsh@2, powershell@2 task selection), (4) MSYS_NO_PATHCONV and MSYS2_ARG_CONV_EXCL workarounds, (5) Windows self-hosted agent configuration, (6) cross-platform script patterns, (7) line-ending issues (CRLF vs LF) in checked-in scripts, (8) Docker on Windows agents, (9) UNC and long-path limitations. Provides: shell-task selection matrix, path-conversion env-var recipes, line-ending fixes, and a Windows-specific pipeline troubleshooting playbook.
Tool use, Agent Skills, and MCP for the Claude API (platform.claude.com). Tool definitions (JSON Schema), tool_use/tool_result, server tools (code execution, web search, web fetch, memory tool, tool search), client toolset (computer use, browser use), Agent Skills (SKILL.md), Skills API GA (/v1/skills), MCP connector, mcp_servers, mcp_tool_listing, MCP tunnels, and Tunnels API.
Create Docker containers for Huawei Ascend NPU development with proper device mappings and volume mounts. Use when setting up Ascend development environments in Docker, running CANN applications in containers, or creating isolated NPU development workspaces. Supports privileged mode (default), basic mode, and full mode with profiling/logging. Auto-detects available NPU devices.
AI autonomous research agent for LLM training optimization using opencode as the agent. The agent autonomously modifies train.py, runs experiments, evaluates val_bpb, and iterates to find the best model. Use when: “run autoresearch”, “start experiment”, “train model”, “autonomous research”, “optimize LLM training”.
Bash 5.3 new features and modern patterns (2025). Proactively activate for: (1) Bash 5.3-specific features (BASH_TRAPSIG, in-shell command substitution, ${|} REPLY syntax), (2) checking Bash version compatibility, (3) migrating scripts to take advantage of 5.3 additions, (4) C23 conformance changes in Bash 5.3, (5) new shopt and bind options, (6) trap handling improvements, (7) wait -p enhancements, (8) READLINE_ARGUMENT and history expansion changes. Provides a complete 5.3 feature reference, version-detection snippets, compatibility shims for older Bash, migration recipes, and POSIX.1-2024 alignment notes
Expert in bridging the gap between business needs and technical solutions. Specializes in Requirements Engineering, BPMN, and Agile User Stories. Use when gathering requirements, creating user stories, modeling business processes, or translating business needs into technical specifications.
Harden a plan with independent Claude/Codex review, then optionally build and cross-inspect it. Start in either Claude Code or Codex: the host plans and the other provider reviews. Use for claudex this plan, claudex-loop, or the legacy crucible trigger; not for trivial edits.
Enforces user-facing data portability in an offline-first app — backup/restore (exact, machine round-trippable) kept strictly separate from export/report (human, lossy, never a restore source); a versioned envelope carrying formatVersion, schemaVersion, appVersion, exportedAtUtc and a payload checksum so a restore refuses a file from a newer app instead of corrupting the database; restore as an all-or-nothing import into a staging database swapped in only after validation; canonical values in machine formats (integer minor units, SI integers, ISO-8601 UTC, stable ids) and never localized numerals; RFC 4180 CSV quoting plus the =/+/-/@ formula-injection escape; streaming writes published by atomic rename; the share sheet behind an injected Gateway; and an export→import→export round-trip test on a hostile fixture. Use when adding or changing export, backup, import, restore, share, or CSV/JSON/PDF output, writing an import validator or merge-vs-replace policy, or handling a picked file.
Creates rigorous, validated models of entities, relationships, and constraints for database schemas (SQL, NoSQL, graph), knowledge graphs, ontologies, API data models, and taxonomies. Covers relational, document, graph, event/time-series, and dimensional schema patterns with lifecycle modeling, soft deletes, polymorphic associations, and hierarchies. Use when user mentions "schema", "data model", "entities", "relationships", "ontology", "knowledge graph", or when data structures need formalization.
Use when the user wants to iterate on an existing generated Power Apps mobile app after /create-mobile-app: update Application Insights configuration, the plan, data model, native capabilities, design, screens, generated app code, and preview without restarting the full project flow. Natural-language triggers include 'the app already exists', 'edit my app', 'change this screen', and 'add a feature to this app'.
Inspect, debug, and automate Chrome directly through the Chrome DevTools Protocol (CDP). Use this skill whenever a task needs a real web browser: open a URL, navigate pages, take a screenshot of a web page, read the DOM or accessibility tree, click buttons, fill and submit forms, type text, wait for elements, scrape page content, evaluate JavaScript on a page, or check console errors and failed network requests. Works against the user's local Chrome or a remote Cloudflare Browser Rendering endpoint over WebSocket. Relevant to browser automation, web scraping, page screenshots, end-to-end UI checks, DevTools inspection, and handing cloud browser control to a human for login, MFA, CAPTCHA, sensitive input, or approval. No Chrome DevTools MCP, Puppeteer, or Playwright required.
Fix Control renderer issues that UI5 linter reports but cannot auto-fix. Use this skill when linter outputs these rules: - `no-deprecated-control-renderer-declaration` - For missing renderer declaration, string-based renderer declaration, implicit renderer auto-discovery (removed in modern UI5) - `no-deprecated-api` - For missing apiVersion:2 in renderer objects, missing IconPool import when using oRm.icon(), deprecated rerender() override. NOTE: For Library.init() apiVersion errors, use fix-library-init instead. - `ui5-class-declaration` - For non-static renderer property in ES6 classes Trigger on: "missing a renderer declaration", "Deprecated declaration of renderer", "deprecated renderer", "apiVersion" (in renderer context), "IconPool", "rerender", "renderer must be a static property" Converts legacy renderer patterns to modern apiVersion: 2 format with proper module imports.
Fix deprecated control/class/interface/type issues that UI5 linter reports but cannot auto-fix. Use this skill when linter outputs: - `no-deprecated-api` with messages about deprecated class, interface, type, property, or property of class - Messages like "Use of deprecated class '...'", "Use of deprecated interface '...'", "Use of deprecated property '...'" Trigger on: deprecated control usage in JS `new` expressions, deprecated properties in constructor settings, deprecated interfaces/types in metadata, deprecated controls/properties in XML views, sap.m.MessagePage modernization to IllustratedMessage. Use this skill whenever you see 'deprecated class', 'deprecated property', 'deprecated interface', or 'deprecated type' in linter output, even if the user doesn't explicitly mention deprecation. Provides replacement guidance for deprecated UI5 controls and their properties.
Production-grade animation patterns for React and Next.js using Motion (formerly Framer Motion). Use this skill whenever the user asks to animate components, add transitions, create scroll-triggered effects, implement page transitions, layout animations, gesture interactions, or any kind of motion/animation in a React or Next.js project. Also trigger when code imports 'framer-motion', 'motion/react', or 'motion/react-client', or when the user mentions: animation, transition, fade-in, slide, parallax, scroll animation, exit animation, AnimatePresence, motion.div, spring, gesture, drag, hover animation, stagger, whileInView, or layout animation.
Upload local images or videos to a GitHub PR and embed them in the description or comments. Use when asked to "attach screenshots to PR", "add images to PR", "upload test results to PR", "embed screenshots in PR description", "add before/after images to PR", "attach UI screenshots", "show test results in PR", "add visual evidence to PR", or any request involving images and PRs. Always use this skill when the user wants to visually document changes in a pull request, even if they don't use the word "upload" — phrases like "put the screenshot in the PR" or "show the image in the PR" should trigger this skill. Prefers GitHub CLI's native `--attach` flag (gh 2.99.0+), which needs no browser at all, and falls back to browser automation (Chrome DevTools MCP, then Playwright MCP, then agent-browser) only when `gh --attach` is unavailable.
Guides architects on when and how to use goal-seeking agents as a design pattern. This skill helps evaluate whether autonomous agents are appropriate for a given problem, how to structure their objectives, integrate with goal_agent_generator, and reference real amplihack examples like AKS SRE automation, CI diagnostics, pre-commit workflows, and fix-agent pattern matching.
Drive ARKit-blendshape-rigged head/face models in Unity with the wearer's facial expressions on Meta VR via Meta Movement SDK (face tracking + A2E). Use when a user has an FBX with the 52 ARKit blendshapes (any prefix, _L/_R suffixes) and wants it to animate from face tracking on Quest Pro / Quest 3 / Quest 3S.
Laravel + Inertia.js + Vue 3 integration patterns. Use when building Inertia page components, handling forms with useForm composable, managing shared data, or implementing persistent layouts. Triggers on tasks involving Inertia.js, page props, form handling, or Laravel Vue integration
Override built-in Markstream node renderers and add trusted custom tags across Vue, React, Solid, Svelte, and Angular. Use when Codex needs scoped `setCustomComponents`, renderer-local React `streamingComponents` or `htmlComponents`, Angular/Svelte `customComponents`, Vue app-scoped components, node override keys such as `image`, `code_block`, `mermaid`, or `link`, or nested renderers for tags such as `thinking`.
Generate Mermaid diagrams (.mmd) and export to PNG/SVG/PDF using mmdc CLI or Kroki API. USE THIS SKILL when user mentions diagram, flowchart, sequence diagram, class diagram, ER diagram, state machine, architecture, visualize, git graph, 画图, 架构图, 流程图, 时序图, 类图, ER图, 甘特图, 状态机. PROACTIVELY USE when explaining ANY system with 3+ components, API flows, authentication sequences, class hierarchies, database schemas, or state machines. Supports 17+ diagram types with fully automatic layout.
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.