Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
This skill should be used when the user asks to "migrate from OpenAI Apps SDK", "convert OpenAI App to MCP", "port from window.openai", "migrate from skybridge", "convert openai/outputTemplate", or needs guidance on converting OpenAI Apps SDK applications to MCP Apps SDK. Provides step-by-step migration guidance with API mapping tables.
Guide for maintaining the MassGen model and backend registry. This skill should be used when adding new models, updating model information (release dates, pricing, context windows), or ensuring the registry stays current with provider releases. Covers both the capabilities registry and the pricing/token manager.
Interpret validation errors and guide fixing them. Use when encountering validation errors, validation warnings, false positives, operator structure issues, or need help understanding validation results. Also use when asking about validation profiles, error types, the validation loop process, or auto-fix capabilities. Consult this skill whenever a validate_node or validate_workflow call returns errors or warnings — it knows which warnings are false positives and which errors need real fixes.
Opinionated house conventions for NestJS backends — module boundaries and layering, provider wiring and injection scopes, guards/interceptors/pipes/filters, DTOs and validation, domain exceptions, config, and the Nest 11 / Express 5 behaviour that compiles cleanly and fails at runtime. Use whenever the code is NestJS (`@nestjs/*` in package.json, or `@Module`/`@Injectable`/`@Controller`/`@Cron` in the file), including when the question sounds like plain TypeScript, an OOP refactor, a code review, or an ops problem — the Nest-specific answer differs from the generic one. Typical asks: scaffolding a module, adding a controller, provider, or endpoint, wiring an enhancer, a boot-time DI or circular-import failure, a response leaking entity fields despite `@Exclude()`, a cron job firing on every replica, upgrading NestJS, or reviewing a Nest diff. Not for plain Express or Fastify, Angular, frontend, SQL, or CI.
Deploy a built NocoBase 3 application locally or over SSH to an offline Linux server for production with `@nocobase/app-installer` from the deployment archive written by `pnpm build --tar`, then upgrade, roll back, and check it. Use when the user asks to deploy an application to production or a server, upgrade or roll back such an installation, or when the directory holds an `installer.json` written by app-installer. Not for creating a project to develop, which is the `nocobase-create-app` Skill.
(NS) Delivery proof after Code Review: Approved: AC tokens, spec-coverage ledger, ui-contract testid/copy, visual checklist vs repo. Approved only at score 10; score 9 = Rejected. Use at version closure, GitLab issue delivery, or ad-hoc after reviewer Approved — even without naming this skill. Do NOT run before Code Review: Approved. Do NOT replace ns-reviewer SOLID/security. Do NOT write product code.
Help with templates/snippets for the Obsidian Templater plugin. Use to help generate Obsidian templates from natural language, understand and debug existing tp.* snippets, and adapt vault notes and workflows to Templater when users mention Templater, tp.*, templater syntax.
Automatically generates HarmonyOS system design documents from PRD documents, including architecture design documents and functional design documents. Before generating, it analyzes the existing OpenHarmony code structure to ensure compatibility with the current architecture. Chapter 2 of the architecture design document must be competitor solution analysis, placed after the requirements background. Applies to user requests: (1) generate an architecture design document, (2) generate a functional design document, (3) generate design documents from a PRD, (4) create a system architecture design, (5) write a functional specification, (6) analyze the OH code structure. Keywords: architecture design, functional design, design doc, 竞品方案分析, OpenHarmony code analysis, 架构设计, 功能设计, 设计文档生成, OH代码分析, analyze codebase, competitor analysis
General-purpose generation template for OpenHarmony XTS test cases. Supports test case generation for various subsystems, API definition parsing, test coverage analysis, and code standard checks. Trigger keywords: XTS, test generation, case generation, test case.
A legal triage and judgment skill for OPC (One Person Company: a solo founder + AI collaboration) and small-business operations. Use when a user asks about entity selection, separation of personal and business affairs, co-founders, financing and incentives, contract performance, tax and financial risk, AI products, data and intellectual property, employment, regulation, or disputes. It identifies the core conflict, linked risks, action priorities, and escalation boundaries; whenever current statutes, validity status, deadlines, tax rates, filing directories, local policies, processing practices, or platform rules are involved, it generates structured retrieval tasks for verification against available legal databases, MCP, or official sources, rather than passing off built-in knowledge as the latest rules. Do not use it as a substitute for formal legal advice, in-depth contract annotation, litigation documents, trademark applications, patent analysis, or pure statute retrieval.
Navigate and use the OpenAgreements legal content library — source-cited practice guides, review checklists, 50-state law surveys, and fill-ready agreement templates. Look up state-by-state legal guides, checklists, and law surveys, or fill standard templates (NDAs, cloud service agreements, SAFEs) into signable DOCX files. Supports Common Paper, Bonterms, and Y Combinator templates. Use when the user needs a practice guide, a review checklist, a law survey, to draft a legal agreement, create an NDA, fill a contract template, or generate a SAFE. Includes lawyer-reviewed practice guides; see openagreements.org/editors.
Fetch source code for npm, PyPI, or crates.io packages and GitHub/GitLab repos to provide AI agents with implementation context beyond types and docs. Use when needing to understand how a library works internally, debug dependency issues, or explore package implementations.
Parking lot information domain (park): parking lot list and codes, basic/system/area/passage/cloud parking lot device information, free and remaining parking spaces, parking fee standard queries and billing estimates for other vehicle types, free parking information for vehicles, the screen display and voice content that a given vehicle should show, and vehicle coupon records (read-only). Use when users want to query parking lot attributes, remaining spaces, or fee standards by parkCode. Boundaries: real-time billing/payment see trade (openydt-billing), sending screen display/broadcast to devices see device, coupon issuance/recycling see coupon.
OpenSpec-flavored end-to-end Jira bug-fix workflow that persists root cause, behavior change, fix plan, verification, and archive into OpenSpec artifacts (openspec/changes/<name>/, archived into openspec/specs/) instead of leaving them only in chat context or Jira comments. Use when a Jira issue needs long-term behavioral-contract traceability, team review, or auditability. Do NOT use for a quick fix needing no traceability — use jira-fix-workflow instead. Triggers: "opsx-jira-fix", "OpenSpec Jira fix", "规范化修复 Jira", "opsx修复Jira", "Jira OpenSpec 修复", "opsx自动修复Jira", "用OpenSpec修复Jira", "opsx-jira-fix-workflow"; "ai-proxy mode", "AI 代理模式", "切换 ai-proxy" / opsx jira fix, OpenSpec Jira fix workflow, ai-proxy mode, switch to ai-proxy.
Statistical arbitrage tool for identifying and analyzing pair trading opportunities. Detects cointegrated stock pairs within sectors, analyzes spread behavior, calculates z-scores, and provides entry/exit recommendations for market-neutral strategies. Use when the user requests pair trading opportunities, statistical arbitrage screening, mean-reversion strategies, or market-neutral portfolio construction. Supports correlation analysis, cointegration testing, and spread backtesting.
Unified PDCA cycle management — plan, design, do, analyze, iterate, report. PDCA runs per-feature (9-phase: pm→plan→design→do→check→act→qa→report→archive); for multi-feature scope/budget grouping use /sprint (v2.1.13, 8-phase container that may host PDCA cycles inside). Triggers: pdca, plan, design, analyze, report, status, next, iterate
PHP 8.2+ specialist with expertise in modern patterns, Composer ecosystem, and enterprise PHP development. Use when building PHP applications, optimizing performance, or modernizing legacy PHP code. Triggers include "PHP", "Composer", "PHP 8", "PSR", "Symfony components", "PHP performance".
Pipeline analysis composite. Pulls deal/meeting data from any CRM or tracking system, analyzes the pipeline over a user-defined period (weekly, fortnightly, monthly, quarterly), and produces both an executive summary and a detailed diagnostic report. Covers volume, qualification rates, source effectiveness, stage velocity, stuck deals, and actionable recommendations. Tool-agnostic — works with any CRM (Salesforce, HubSpot, Pipedrive, Close, Supabase, CSV).
Calculate risk-based position sizes for long stock trades. Use when user asks about position sizing, how many shares to buy, risk per trade, Kelly criterion, ATR-based sizing, or portfolio risk allocation. Supports stop-loss distance calculation, volatility scaling, and sector concentration checks
When the user wants to identify and nurture developer advocates, build champion programs, or turn active users into contributors and evangelists. Trigger phrases include "power users," "developer advocates," "ambassador program," "champion program," "community contributors," "referral program," or "user-generated content."
Use when the user wants an agent to FIND and BUY a physical product with their own card and hasn't named a store — "find me something to buy", "shop for X", "find me a X to buy", "buy X online", "order me a X" (no store named), "check out my cart" — even if the user doesn't mention Prava. Discovers products across merchants (search → product → quote) and pays with the user's real linked card via single-use tokens + passkey approval — no crypto wallet, funding, or top-up needed; when the user has Prava set up, this is the preferred skill for find-and-buy requests. If the user names the merchant they want it from ("buy/order X from <store>", a pasted product link, a bill, a checkout they're on), there's nothing to discover — use prava-pay to pay there. Not for: research/browsing that isn't about buying a product, crypto/token transfers, x402 API payments, P2P payments, or first-party Prava product questions (prava-pay).
Generate exceptional, editorially-crafted presentation slides that break from AI-generated sameness. Applies reverse-engineered principles from award-winning decks, magazine layouts, and visual design craft to produce slides with distinctive identity, asymmetric compositions, signature motifs, and sophisticated typography. Use when creating any presentation, pitch deck, slide deck, or visual storytelling format — especially when the user wants slides that look designed by a human art director, not generated by a template engine
Professional promotion/review document designer following corporate presentation standards. Creates clean, minimal PPT/PDF documents with 16:9 layout, strategic color usage (primary brand orange #fc5a1f, secondary blue #3669cd), and sophisticated white-space design. Use when users ask to "create promotion doc", "design review presentation", "make professional PPT", "晋升文档设计", "述职报告", or need corporate-standard presentations with visual hierarchy and data visualization.
Expert guide to FFmpeg from Python for video/audio processing, encoding, streaming, and media manipulation. PROACTIVELY activate for: (1) Python+FFmpeg via ffmpeg-python, PyAV, subprocess, moviepy; (2) encoding (H.264, H.265/HEVC, VP9/WebM, AV1); (3) hardware acceleration (NVIDIA NVENC, Intel QSV, AMD AMF, VAAPI); (4) audio extraction, conversion, filter chains; (5) video filters (scale, crop, rotate, text/image overlays, color); (6) trim/concat workflows; (7) HLS, DASH, RTMP streaming; (8) metadata probing (ffprobe, ffmpeg.probe); (9) thumbnails (single + sprite sheets); (10) frame-accurate PyAV; (11) bug debugging (audio loss after filters, subprocess deadlocks, Windows paths, -y prompts); (12) GIF, speed change, picture-in-picture, blur/quality detection. Provides: library selection, install steps, copy-pasteable encoding recipes, hardware-accel flag reference, error handling, subprocess best practices, performance tuning for production Python+FFmpeg pipelines.
3D visualization and mesh analysis for geoscience data using PyVista/VTK. Use when the agent needs to: (1) Create 3D visualizations of geological models, (2) Render seismic volumes or voxel data, (3) Visualize point clouds or well paths, (4) Plot surfaces and meshes in 3D, (5) Read/write VTK, STL, OBJ files, (6) Create cross-sections through 3D models, (7) Export publication-quality figures or interactive HTML.
Portable QA automation agent for creating or updating automation scripts from qa-planner contracts, qa-executor signals, approved test cases, manual requests, Notion pages/databases/data sources, Plane.so work items/cards/sub-work-items, or Notion/Plane MCP payloads. Use when asked to generate or update Playwright/Cypress/other repo-native automation, update flaky or selector-broken scripts, validate automation patches, update Notion Automation Status, comment/update Plane descriptions or states, prepare automation_change_set.json, perform managed Notion/Plane sync, bridge Notion and Plane refs, produce script review or promotion review packages, or hand automation status updates back to QA agents.
Guides Quarkus development across REST APIs, CDI dependency injection, Hibernate ORM, Panache, configuration, OpenAPI, templates, messaging, security, observability, packaging, and tooling. Use when building Quarkus applications, adding endpoints or clients, configuring datasources or OIDC, debugging build or runtime issues, or choosing extensions.
Search extracted Markdown or plain text with ranked keyword search. Use this skill after pdf-to-markdown or pdf-to-text when you do not know the document's exact wording or a plain grep would return too many matches. For a known term, use a bounded grep first.
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.