Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Create and evolve UI specifications mocked up in HTML + Tailwind CSS: annotated desktop and mobile screens, flows, functional rules, open questions, and an iteration log, for iterating on an app or tool before development. If the interface has not yet been approved, first propose three options that differ in layout (navigation, screen organization) and theme (colors, typography, shapes), to choose from or combine. Can also start from an existing site or code to reuse its style and specify new features. Use whenever someone mentions mockups, wireframes, UI specs, functional specifications with screens, an interface prototype, flows to mock up, a visual style guide or direction for an app, or a new feature to mock up on an existing site, even without saying “specs.” Also use during development to keep specs up to date with the implementation (screen status, differences).
Use when building B2B or wholesale commerce on Spree 6 — the end-to-end "build a B2B store" setup, companies/buyer organizations (company trees, divisions, members, invitations, company address book, buying on a company's behalf), catalogs and negotiated pricing (assortments, owned price lists, percentage adjustments, catalog assignment to companies or customer groups, channel default catalogs), customer groups, gated storefronts (`storefront_access` login_required / prices_hidden), the Next.js wholesale portal, minimum order quantity / order multiples / order minimums, PO numbers, and orders on account (pay by invoice, staff-keyed draft orders with negotiated prices, `payment_pending`), and freight shipping (cartons, pallets, unpriced freight rates). Common phrasings include "B2B", "wholesale", "build a B2B store", "pay by invoice", "net terms", "negotiated line price", "draft order for a company", "trade customers", "company account", "Spree::Company", "division", "buyer organization", "negotiated prices", "customer-specific pricing", "catalog", "price list per company", "MOQ", "minimum order quantity", "order multiple", "case pack", "order minimum", "PO number", "purchase order reference", "login to see prices", "members-only store", "wholesale portal", "X-Spree-Channel", "freight", "pallet", "carton".
Use when building or extending a multi-vendor marketplace on Spree 6 — sellers/vendors, seller onboarding requirements, seller product review (submit/approve/reject), split checkout into order groups, commissions (rates, rules, commission lines, commission tax), seller payouts and the transfer/payout ledger, Stripe Connect, a custom payout provider, the seller panel (`@spree/seller-dashboard`), or the Seller API (`/api/v3/seller`, `@spree/seller-sdk`). Common phrasings include "marketplace", "multi-vendor", "vendor", "seller", "Spree::Seller", "invite/approve a seller", "onboarding checklist", "seller requirements", "order group", "isOrderGroup", "split order per seller", "commission rate", "commission line", "seller payout", "payout provider", "SweepDueJob", "Stripe Connect", "seller API", "X-Spree-Seller-Id", "seller dashboard"
Use when connecting Spree 6 to an external system through a provider contract — a tax engine, carrier rate shopping, a WMS/3PL or label provider, a payment gateway, a search engine (Meilisearch, Typesense, Algolia), seller payouts, digital downloads/license keys, SSO/OpenID Connect or a custom JWT issuer, OpenTelemetry/APM, or an ERP/PIM/DAM. Common phrasings - “custom tax provider”, “Avalara”, “delivery rate provider”, “EasyPost”, “fulfillment provider”, “custom payment method”, “search provider”, “payout provider”, “digital asset provider”, “SSO”, “Okta/Entra login”, “Spree::Integration”, “store credentials for a provider”, “Spree.integrations”, “which registry do I append to”. A map of every plug-in point - base class, registration call, where it is selected, and the doc to read
Test-Driven Repair — given a failing test, spawn a bounded headless `claude -p` (Read/Edit/Bash only) that makes the test pass without modifying it. Modeled on agent-harness-generator's ADR-175 Test-Driven Repair mode. Bounded cost via --max-budget-usd, bounded capability via --allowedTools. Closes the loop the TDD plugins didn't — we generate tests, this fixes the code to satisfy them.
Tiger Brokers OpenAPI Java SDK — complete skills for AI coding tools. Covers SDK setup, market data queries, stock/futures/options trading, real-time push subscriptions, account management, and strategy examples. Use when building Java trading applications, querying market data, placing orders, or integrating Tiger Brokers API. Tiger Brokers OpenAPI Java SDK complete skill set. Covers SDK setup, market data queries, stock/futures/options trading, real-time push subscriptions, account management, and strategy examples. For building Java trading applications, querying market data, placing trades, or integrating the Tiger API. Auto-activate on these keywords: 行情、报价、价格、K线、快照、买卖盘、深度、买入、卖出、下单、撤单、改单、交易、持仓、资金、账户、订单、委托、期权、期权链、到期日、期货、推送、订阅、选股、筛选、tigeropen、tiger API、Java SDK、quote、price、K-line、order、buy、sell、trade、position、asset、account、option、future、push、scanner
Apply whenever you write or review a command that decides whether work is done: a test invocation, an acceptance check, a CI gate, a shell assertion, a grep-based verify step, or an upgrade, deployment, migration, or install whose success you will report. Names the failure signal, grounds paths, and catches the shapes that pass green while measuring nothing.
Deutsche Texte natürlich, sachlich und leserorientiert schreiben oder überarbeiten. Nutze den Skill für Artikel, E-Mails, Berichte und Social-Media-Texte sowie zum Vermenschlichen vorhandener Entwürfe. Erhält Fakten und persönliche Stimme, arbeitet mit einem Überarbeitungsablauf von Relevanz und Substanz bis zu Artefakten, reduziert schematische Sprachmuster (Musterkatalog mit Prüfdatum) und unterscheidet Stilprüfung, KI-Detektoren, Textwasserzeichen und C2PA-Herkunftskennzeichen. Keine Garantie für Nichterkennbarkeit oder Wasserzeichenfreiheit.
Build and run workflow for WinUI 3 apps with WinApp CLI 0.7+ — project creation with winapp new, per-app NuGet analyzer setup, project-mode winapp run, Native AOT publish runs, crash diagnosis, and prerequisites. Use when creating, building, running, or fixing build errors in a WinUI 3 project
Automated UI testing for Windows desktop apps — generate a batch test script with the `winapp ui` UI Automation harness (WinApp CLI 0.7+), run all tests in one pass, read results. Covers assertions, interactions, keyboard/touch/pen input, file pickers, dialogs, persistence, accessibility, and screenshot/video capture for Win32, WPF, WinForms, and WinUI 3, in Windows Sandbox or locally.
Guide safe Wuji Glove calibration with the wuji CLI. Use when the user asks to run or monitor `wuji calib hand-model` or `wuji calib tactile`, interpret calibration progress or errors, cancel an active calibration, inspect the resulting hand model, or choose between hand model and tactile calibration. Route each mode without applying the hand model's user-and-hand rules to tactile's device-specific model.
Audit the choices an implementing agent made, not its diff — a pure decision audit that traces the session's history into a choices ledger, changes no code, and never blocks an unsupervised run. Working code still embeds architecture the user never chose; surface it because future work inherits it. Use when the user wants to review the decisions the AI made on their behalf, before merging or committing AI-implemented work, when integrating a delegated subagent's pass, or when a fix "works" but might be a point fix.
Automated publishing skill for WeChat Channels. Use when the user wants to publish a video to WeChat Channels. Includes features such as retrieving the login Cookie, uploading videos, setting titles and topics, scheduling posts, and declaring original content.
Expert knowledge for Azure Application Gateway development, including troubleshooting, best practices, decision-making, architecture and design patterns, limits and quotas, security, configuration, integrations, coding patterns, and deployment. Use when configuring App Gateway/AGIC for AKS, WAF/TLS, Private Link, header rewrites, or autoscaling, and for other Azure Application Gateway development tasks. Not for Azure Load Balancer (use azure-load-balancer), Azure Front Door (use azure-front-door), Azure Web Application Firewall (use azure-web-application-firewall), or Azure Virtual Network (use azure-virtual-network).
Turn a one-line motion / animation / design brief into the full prompt: a committed look grounded in five reference images, three motion references, numbered beats with a turn, and the refs and assets attached. Also tightens edit notes. Use for \"/blank-page\", \"expand this prompt\", \"make this prompt more specific\", \"what is missing from this prompt\", \"help me write the prompt for X\", or whenever a one-line animation / composition / design brief is about to be handed to an agent.
Helps collaboratively clarify a task, test an idea, find the root cause of a problem, choose an approach, or discuss someone else's review comments without implementing anything. Use when the user asks to discuss a task, investigate a problem, compare options, decide which PR or MR comments to address, or asks “What do you think?” or “What’s the best way to do this?” Do not use when the user explicitly asks for a plan, implementation, a code review with comments, or project knowledge to be saved
Communications-domain literature review with Claude-style knowledge-base-first retrieval. Use when the task is about communications, wireless, networking, satellite/NTN, Wi-Fi, cellular, transport protocols, congestion control, routing, scheduling, MAC/PHY, rate adaptation, channel estimation, beamforming, or communication-system research and the user wants papers, related work, a survey, or a landscape summary.
Choosing or implementing an algorithm or data structure — asymptotic complexity, data-structure selection, sorting & searching, dynamic programming, graph algorithms, divide & conquer, greedy algorithms, string/sequence algorithms, and the at-scale toolbox (Bloom filters, HyperLogLog, Count-Min Sketch, reservoir sampling, consistent hashing, external merge sort, Aho-Corasick, MinHash/LSH). Trigger on tasks involving "what's the right algorithm for…", performance-critical code, code with nested loops over the same input, recursive solutions, shortest-path / scheduling / matching / DP problems, code review for accidental O(n²) blowup, and any "how do I do X at scale / on a stream / without enough RAM" question — even if the user doesn't explicitly mention "algorithm" or "complexity."
Draft marketing content across channels — blog posts, social media, email newsletters, landing pages, press releases, and case studies. Use when writing any marketing content, when you need channel-specific formatting, SEO-optimized copy, headline options, or calls to action.
Fixes content and semantic HTML issues. Adds server-side rendering, a correct heading hierarchy, semantic landmark elements, ARIA labels, image alt text, the lang attribute, descriptive link texts, form labels, autocomplete tokens, keyboard access, document titles, button names, correct aria-hidden use and widget state so AI agents can read and operate a page through the HTML and the accessibility tree. Use when asked to "fix semantic HTML", "add SSR", "fix heading hierarchy", "add alt text", "improve semantics score", "fix accessibility for AI", "fix form labels", "add autocomplete", "fix keyboard navigation", "fix button names", "fix aria-hidden", "add aria-expanded", or any semantic HTML task.
Review contracts against your organization's negotiation playbook, flagging deviations and generating redline suggestions. Use when reviewing vendor contracts, customer agreements, or any commercial agreement where you need clause-by-clause analysis against standard positions.
Critique an existing design (web page, HTML/React UI, screenshot, generated image, or Figma frame) with a repeatable procedure: capture it (Playwright screenshots or Read for images), run automated checks (axe, style inventory), score against a weighted rubric, and output severity-ranked fixes with code, then re-verify. Use after any design is produced or when asked to review one. Trigger phrases: "design critique", "review design", "improve design", "design audit", "design feedback", "evaluate UI", "design score", "accessibility audit", "visual review", "design assessment"
Find candidate stocks with steadily growing dividends, stable operating quality, and valuations that have returned to a reasonable range. Output entry watch zones, growth support, and invalidation boundaries. Suitable for selecting stocks for both income and growth, long-term allocation, and buying on pullbacks.
Check documentation impact before completing a change or declaring it ready to ship. Use when a change may affect user, operator, or developer instructions, when asked whether docs need updates, or for a documentation pass. Supports author and read-only review modes without requiring a PR or spec.
Enforces a typed-error spine — a hand-rolled sealed Result<T,F> plus one per-boundary sealed Failure carrying a stable code and typed params (never a localized string), returned instead of thrown; recoverable failures are values, only bugs throw. Convert-at-boundary catches narrowly with an on-clause and logs the original error+stack BEFORE returning a typed Failure; call sites switch exhaustively with no default:; the taxonomy of what the global error net (FlutterError.onError + PlatformDispatcher.onError, installed by app-startup-and-bootstrap) routes into, plus Isolate.run re-wrapping; mechanism selection (Error vs assert vs Exception vs sealed outcome) and @useResult; and a never-lose-data layer (one-transaction-per-mutation, debounced autosave drafts, optimistic soft-delete/Undo). Use when writing result.dart/failures.dart, a try/catch or sealed-switch default:, wiring bootstrap handlers, a repository/service/DAO boundary, or transactions, drafts, or soft-delete/Undo.
Eva Preflight 2.5.0 entry point for pre-publication review. Use when the user explicitly asks for a pre-publication check, final-draft review, whether a piece is ready to publish, a full review before posting, or any major issues in a draft. Review natural-language content that the user or team has substantially completed but not yet published. Depending on whether it is a short video, article, general social post, or sponsored content, read only the existing title, opening, logic, facts, messaging assets, and Brief source of truth; silently check for significant shifts in audience or user problem between the title, opening, and body; then provide one of three publication verdicts and one next step. If the draft explicitly aims to acquire customers or includes product promises or consultation/purchase calls to action, read-only checks of alignment across audience, qualification, content, and product capabilities may be performed as needed. Do not use for a topic, source materials, or outline alone; send published content to eva-review; do not use for action plans, code, product releases, or deployment acceptance.
Score, evaluate, and iteratively improve any content or strategy using an auto-assembled panel of domain experts. Handles copy, sequences, landing pages, strategy docs, titles, charts, recruiting evaluations, or anything else that needs a quality gate. Recursively iterates until all scores hit 90+ (max 3 rounds). Use when asked to: "expert panel this", "score this", "rate these variants", "quality check this", "panel review", "which version is better", "expert score", "evaluate this copy/strategy/page", or when another skill needs a quality gate on its output. Also triggers on: "score this landing page", "expert panel these email variants", "rate this headline", "panel these charts".
Explore PostHog MCP intent clusters — agent goals grouped by semantic similarity, with each cluster's tool distribution and error rates, plus the tool-centric pivot (capture rate per intent, discovery rate against the advertised catalog, description fit, tool overlaps). Use when the user asks "what are agents trying to do with the MCP?", "group the intents", "which goals fail most?", "what does each cluster route to?", "when agents have this intent do they find my tool?", "which tools get mixed up?", wants to recompute the clustering, or pastes an MCP analytics intent-clustering URL.
Investigate individual PostHog MCP sessions — the sequence of tool calls a single agent made in one run, what it was trying to do, and where it went wrong. Use when the user asks "what did this MCP session do?", "show me the tool calls for session X", "what was the agent's goal?", "which sessions had errors?", "who is connecting to my MCP?", or pastes an MCP analytics sessions URL.
Investigate the quality of PostHog MCP tool calls — error rates, latency, reach, and which tools are failing or slow. Use when the user asks "which MCP tool has the highest error rate?", "what's the slowest tool?", "which tools fail most often?", "how reliable is tool X?", wants a tool-quality matrix, or pastes an MCP analytics tool-quality / dashboard URL and asks what it shows.
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.