Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Manage daily operations for Israeli freelancers (osek murshe, osek patur) - invoice aging with collection reminders, utility bill collection via browser automation, tax deadline alerts (VAT, Bituach Leumi, mkdamot, annual report), osek patur threshold monitoring, and organized accountant packages (havila). Use when a freelancer needs help tracking invoices, preparing documents for their accountant, monitoring their osek patur revenue ceiling, or staying on top of Israeli tax filing deadlines. Prevents missed VAT filings (which trigger automatic penalties), forgotten invoice follow-ups, and disorganized handoffs to accountants. Do NOT use for VAT return preparation (use israeli-vat-reporting), e-invoice generation (use israeli-e-invoice), or payroll/employee management.
Grid systems, visual hierarchy, whitespace, and modern CSS layout patterns (grid, flexbox, bento, container queries) for web pages and UI. Use when arranging content on a page or screen. For spacing/radius tokens themselves use design-system; for slide layouts use presentation-design. Trigger phrases: "layout", "grid", "bento grid", "visual hierarchy", "page layout", "flexbox layout", "whitespace", "responsive layout", "composition", "css grid", "container queries".
Connect Claude Code or Codex itself to Logfire telemetry, or route a coding agent's model requests through Pydantic AI Gateway. Use for coding-session recording, capture policies, Gateway agent configuration, temporary proxy launches, status, or removal. Not for instrumenting an application or routing an application's model client.
Query @lynx-js/css-defines compat_data to check Lynx CSS property and nested feature/value support by rendering backend and Lynx version, or check whether a newer dataset version exists. Use when asked whether CSS is supported in Lynx or ReactLynx, which backends support it, when support was added, whether a style change is compatible, whether the bundled data is current, or when inspecting the raw definition JSON.
Refresh the user's competitive and market context. Re-scan named rivals and the category for changes since the brief was written (rebrands, repositioning, new entrants, pricing shifts, renamed categories), and update docs/gtm-cofounder/founder-brief.md and the roadmap with anything that affects positioning. Use on a cadence, or whenever a competitor rebrands, a new tool appears, or the user senses the market has shifted
MCP server building, advanced patterns, and security hardening. Use when building MCP servers, implementing tool handlers, choosing a transport, adding OAuth authentication, wiring MCP Apps UI with @mcp-ui, hardening MCP security, or debugging MCP integrations.
Migrate Python LangChain Deep Agents applications to Pydantic AI and Pydantic AI Harness. Use when the source uses the upstream `deepagents` package or demonstrably reproduces its middleware, backends, skills, memory, subagents, or sandbox contracts. Use `migrating-langchain-to-pydantic-ai` for plain LangChain, LangGraph, or LCEL migrations and `pydantic-ai-harness` for greenfield Harness usage.
Enforces Effective-Dart casing (UpperCamelCase types, lowerCamelCase members/constants, lowercase_with_underscores files) plus architectural role suffixes so a name or grep reveals the layer — Screen/Notifier/Repository/Dao/Service/Gateway/Failure, file=primary-declaration, units-and-semantics in identifiers, booleans as is/has/can/should assertions, no get-prefix, no Hungarian, no SCREAMING_CAPS, grouped-and-sorted imports. Use when creating a file, naming a class/enum/mixin/extension/typedef/variable/function/getter/constant/parameter, organizing imports, choosing a role suffix, or reviewing a diff for naming and directive ordering.
Generate images with Google's Nano Banana Pro (Gemini 3 Pro Image). Use when generating AI images via Gemini API, creating professional visuals, or building image generation features. Triggers on Nano Banana Pro, Gemini 3 Pro Image, gemini-3-pro-image-preview, Google image generation.
Use this skill when the user asks to consult Codex/GPT for planning or code review, or wants a second opinion on code. Trigger phrases include "ask Codex", "consult Codex", "second opinion on code", "consult the oracle". Not for implementation tasks.
Transforms cold outreach from spam into revenue using Hormozi value-first principles. Triggers when a user needs to improve cold emails or DMs, increase response rates, diagnose low-converting outreach sequences, or build a prospecting strategy.
Technical architecture role for Phaser 4 mobile 2D games. Use when defining or reviewing module responsibilities, public contracts, state ownership, and dependency direction, or establishing technical baselines for TypeScript, Vite, Capacitor, scenes, assets, saves, platform adaptation, and reproducible builds.
Create and manage persistent markdown planning files for structured task execution. Use when the user asks to “create a plan”, “track progress”, “start a research project”, or when a task requires more than 5 tool calls and needs structured phase tracking to stay focused and avoid goal drift.
Software implementation planning with optional file-based persistence. Use when asked to plan, when unresolved architecture or scope decisions need a durable record, or when multi-phase implementation needs recovery state. For the full research-and-issue workflow, use the ia-plan command (/ia-plan in Claude Code).
Expert PR and media relations guidance for earned media, press coverage, and reputation building. Use when writing press releases, crafting media pitches, developing journalist relationships, planning embargo strategies, managing crisis communication, placing thought leadership, building analyst relations, submitting award applications, or measuring PR coverage. Use for media outreach, press kits, speaking opportunities, and earned media strategy.
Generate structured presentation outlines with bold statement slides, section dividers, and clear narrative arcs. Use when starting a new presentation, planning a deck structure, or asking "outline a presentation about...", "structure a deck for...", or "create a presentation flow for...". Outputs markdown outlines ready to translate into slides.
Generate action recommendations to scale out, continue holding, or reassess when a stock price approaches, reaches, or crosses a target price. Applies to predefined plan triggers, take-profit discipline execution, and target price reassessment scenarios.
Prompt the user with pointed questions to extract what is in their head about a project — remaining work, what is being avoided, what really matters, what does not. Use when the user says "prompt me", "ask me questions", or wants the agent to figure out priorities by questioning them
Multi-agent QA review team for code changes. Use when the user asks to “review my code”, “run QA”, “qa-team”, “review this branch”, “code review”, “check my changes”, or requests a comprehensive multi-perspective code review of the current branch's changes. Spawns parallel specialist agents (security, database, reliability, compatibility, data integrity, performance, frontend, copy) that independently review the diff and produce a converged report. Also includes two generalist reviewers for convergence validation
Use for Lovrabet development through the rabetbase CLI: create or bind projects (including creating an AppCode project in the current folder), pull Dataset/API facts and maintain SDK clients, or manage datasets, Instant API access policies, database connections, pages, host frontend integration, frontend subapp resources, SQL, Backend Functions, approval flows and workflows, menus, notifications, knowledge bases and kb search, Agent Context Rules, roles, files, OCR, deployment metadata, platform issues, and 钉钉沙箱账号绑定. Trigger when the user mentions rabetbase, Lovrabet development, AppCode, Dataset, Instant API allow/deny/route policy, project creation, api pull/codegen, dblink, page, host frontend integration, subapp resource versions, SQL, BFF, creating or designing an approval flow, workflow, business process, or multi-step approval, menu, notification, knowledge base, kb search, Agent Context Rules, RULES.md, DATABASE.md, rule list/get/set, role, file/OCR, or related development workflows
Detect React and TypeScript maintainability smells across the whole component, hook, and module, explain the cost of each, and propose a fix with a source link. Covers prop and API design (naming, booleans, callbacks, variants, controlled state, generics, refs, styling, JSDoc), state and data flow (redundant state, stale closures, context and ref misuse), effects and lifecycle, component structure and hooks, rendering correctness (keys, mutation, impure renders, hydration, the leaked 0 from &&), accessibility and focus management, async handlers, error boundaries, unsafe HTML, TypeScript discipline (any, casts, untyped catch, exhaustiveness), and a cross-file duplication pass. Defers memoization and re-render performance to react-compiler and color literals to theme-colors. Use when reviewing or writing React components or hooks, auditing for maintainability, or asked to sniff, smell-check, lint, or clean up a codebase, folder, file, or pasted snippet. Respects native HTML attributes and library conventions.
Read reviewed Word/WPS docx files and extract comments and track changes into a markdown reconciliation table or JSON for item-by-item decisions. Use when a reviewed contract, agreement, manuscript, or report is returned and feedback must be read; for “extract docx comments,” “reconcile review feedback,” “read the revisions,” “who commented on what and where”; for reviewed drafts exported from WPS Cloud Docs/Word Online collaboration; when a returned file has tracked changes and you need to see what the other party changed; or when comparing a clean, unannotated and unrevised document against a known baseline. Revision-aware: bare python-docx reads miss inserted text (w:ins) and fail to recognize whole-paragraph deletions (w:del). This skill uses the OpenXML SDK engine to read supported comments, paragraph additions/deletions, and text views; see the main documentation for scope. Read-only; does not modify the other party’s file. Out of scope: generating/formatting docx → docx-creator or minimax-docx; PDF comments are not supported.
Analyze and optimize Roblox scene and script performance. Use for mobile frame rate, memory, instance count, rendering cost, physics, particles, lighting, asset budgets, streaming, leaks, and using the rbx-scene-analysis workflow.
Stands up one navigable feature the same way every time — a fixed lib/features/<feature>/presentation/ folder (dumb View + one 1:1 Notifier/AsyncNotifier ViewModel + widgets/ + scoped <feature>_providers.dart), downward-only dependencies with no cross-feature imports, reactive reads as scoped StreamProviders over a repository .watch(), every mutation through one repository single-write-path method, typed Result/Failure surfaced as AsyncValue, a typed go_router route carrying identity in path params, ARB parity across locales, family+autoDispose keying, and EdgeInsetsDirectional-only geometry. Use when adding a screen, tab, or feature folder; wiring a feature Notifier/ViewModel or a scoped stream provider; registering a route; adding ARB keys for a new screen; splitting an over-grown feature; or adding a persisted record its View reads.
Expert-level Terraform infrastructure as code, modules, state management, and production best practices. Use when the user mentions infrastructure as code, devops, or automation, or when the task involves Terraform Basics, Resource Management, Modules, or State Management.
Tiger Brokers OpenAPI TypeScript SDK — complete skills for AI coding tools. Covers SDK setup, market data queries, trading, and real-time push subscriptions. Use when building trading applications in TypeScript/Node.js, querying market data, placing orders, or integrating Tiger Brokers API. 老虎证券 OpenAPI TypeScript SDK 完整技能集。涵盖 SDK 配置、行情查询、交易、实时推送订阅。适用于用 TypeScript/Node.js 构建交易应用、查询行情数据、下单交易。 用户提到以下关键词时自动使用 / Auto-activate on these keywords: 行情、报价、K线、买入、卖出、下单、撤单、交易、持仓、资金、账户、订单、期权、期货、推送、订阅、TypeScript SDK、tigeropen typescript、quote、price、K-line、order、buy、sell、trade、position、asset、option、future、push
[user/auto] Use when structuring output so the key points, actionable steps, and confirmed completion status are easy to find in answers and work results. It does not handle session state restoration, handoff write-ups, task execution, medical advice, or skill maintenance.
A Tushare data research skill for natural-language requests in Chinese. It turns requests such as "how has this stock been doing lately", "help me check the earnings trend", "which sector is strongest recently", "what is northbound capital buying", "export a set of market data for me" into executable workflows for data retrieval, cleaning, comparison, filtering, export and brief analysis. Suitable for research on A-shares, indices, ETFs/funds, financials, valuations, capital flows, announcements and news, sector concepts and macroeconomic data.
Enforces a pure-Dart value-object core that stores every quantity canonically — money as integer minor units keyed to each currency's real ISO-4217 exponent (never *100), physical amounts as SI whole units, timestamps as UTC — and converts only at the presentation edge; forbids double/num money, cross-currency arithmetic, and defaulting an unknown currency to two decimals; routes every division of money through one largest-remainder allocate() primitive so parts always sum to the whole to the exact minor unit; derives totals instead of storing them, links entities by stable id, and injects a Clock instead of DateTime.now. Use when defining or changing Money, Currency, or a unit value object; parsing or formatting an amount; splitting, prorating, discounting, tax/tip, or distributing money; adding a currency or FX rate; converting quantities; or fixing float-money, hardcoded-100, cross-currency, off-by-a-cent, or stored-total-drift bugs.
Use when designing or reviewing Android UI — applies Material Design 3 UX principles covering touch targets, spacing, navigation patterns, accessibility, animation timing, and platform conventions. Includes an M3 compliance audit that scores screens across 10 categories. Complements the compose skill with design-level decisions.
Convert raster drawing references—including screenshots, scans, photos, sketches, exported PNG/JPG files, and photographed plans—into structured, editable AutoCAD DWG/DXF drawings with explicit evidence tracking, input preflight, specification validation, output inspection, and visual comparison. Use for dimension-driven redraws, hybrid reconstruction, visual tracing, or geometry-only redraws when the source is an image rather than a trustworthy source DWG.
Expert knowledge for Azure Resource Graph development, including troubleshooting, best practices, decision-making, limits and quotas, configuration, integrations, and coding patterns. Use when querying via CLI/PowerShell/REST, using GET/LIST vs. Query, handling paging and quotas, or creating Power BI ARG reports, and for other Azure Resource Graph development tasks. Not for Azure Monitor (use azure-monitor), Azure Policy (use azure-policy), Cost Management (use azure-cost-management), or Azure Resource Manager (use azure-resource-manager).
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.