发现 Skills

Work with the Ticks issue tracker and the Ticker AI agent runner. Use this Skill when managing tasks or issues with tk commands, running AI agents on epics, creating ticks from a SPEC.md, or working in a repository that has a .tick directory. It triggers on phrases like "create ticks", "tk", "run ticker", "epic", "close the task", "plan this", and "break this down."

Runbook for diagnosing and resolving website downtime, server issues, and deployment failures. Quick triage steps for Perry's client sites.

Containerization for TypeScript/Node.js applications. Use when deploying Node.js backends, need consistent dev environments, or setting up CI/CD pipelines. Covers multi-stage builds, docker-compose for development, and production optimization. Choose this skill for containerizing tRPC/Express APIs with Prisma.

Assess a codebase's readiness for autonomous agent development and provide tailored recommendations. Use this when asked to evaluate how well a project supports unattended agent execution, assess development practices for agent autonomy, audit infrastructure for agent reliability, or improve a codebase for autonomous agent workflows. Triggers on requests like "assess this project for agent readiness", "how autonomous-ready is this codebase", "evaluate agent infrastructure", or "improve development practices for agents".

Guidance for creating and organizing custom sub-agents in local repos, including folder conventions, per-agent structure, and AGENTS.md indexing. Use when asked where to store sub-agents or how to document them.

Complete feature planning from specification to implementation. Create structured specs with user stories and acceptance criteria, then generate detailed implementation plans with TDD workflow, exact file paths, and complete code examples.

Coordinate infrastructure for inventory, stock movement, BOMs, valuation, and multi-location controls while referencing the existing ERP implementation, inventory docs, and the small-business bookkeeping playbook.

Generate end-user manuals and reference guides for ERP modules. Use when the user asks to document a feature, write a user manual, or sync a reference guide. This skill is explicitly separate from doc-architect (which manages AI guidance docs like AGENTS.md).

Build SAAS platforms with pluggable business modules (Advanced Inventory, Restaurant, Pharmacy, etc.) that can be enabled/disabled per tenant without breaking the system. Use when designing modular SAAS features, implementing module toggles, ensuring module independence, and building scalable multi-tenant platforms with optional features.

Production-grade multi-tenant SaaS platform architecture with three-panel separation, zero-trust security, strict tenant isolation, and comprehensive audit trails. Use for designing multi-tenant systems, implementing tenant-scoped permissions, ensuring data isolation, and building scalable SaaS platforms.

Master skill for orchestrating multi-step workflows. Use when generating code for complex processes, agent coordination, or system design. Ensures proper step definition, dependency tracking, error handling, and validation.

Design POS, checkout, and sales-entry web UIs that are simple, accessible, and fast for all ages while integrating all backend actions strictly through APIs. Use for creating or reviewing UI patterns, layouts, components, and workflows for web-based sales recording systems.

Guided interview to create comprehensive project requirements documentation (requirements.md, business-rules.md, user-types.md, workflows.md) for a new SaaS project. Use before bootstrapping the SaaS Seeder Template.

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

Edit PDFs with natural-language instructions using the nano-pdf CLI.

Production-grade Playwright in TypeScript: Page Object Model, fixtures, auto-waiting, user-facing locators, parallel execution, CI integration, visual testing, accessibility. Includes explicit "do not" list for AI agents and 2025-2026 feature awareness. Use when: "Playwright," "browser testing," "E2E test," "end-to-end," "page object." Related: visual-testing, ci-cd-integration, api-testing, test-reliability, accessibility-testing.

Guide for spec-based agent-driven development using structured requirements (EARS notation), technical design documentation, and implementation planning. Use when users want to build features using a specification-first approach, need to document requirements formally, want to generate implementation plans from specs, or when working on projects that benefit from clear requirement-to-code traceability.

>-

>-

>-

Complete guide to packaging AI agents and domain knowledge as Claude Code plugins. Use when the user mentions: Claude Code plugin, claude plugin, plugin packaging, plugin structure, plugin.json, CLAUDE.md plugin, skill file, command file, slash command packaging, MCP plugin, plugin distribution, plugin marketplace, plugin install, plugin directory, plugin testing, plugin versioning, extend Claude Code, Claude Code extension, bundle MCP server, domain expertise plugin

Complete guide to packaging AI agents as Custom GPTs for the ChatGPT GPT Store. Use when the user mentions: Custom GPT, GPT Store, GPT Builder, GPT Actions, GPT packaging, ChatGPT plugin, GPT knowledge files, GPT system instructions, build a GPT, create a GPT, publish GPT, GPT marketplace, GPT distribution, OpenAI GPT, GPT configuration, conversational agent, GPT monetization

Jakob Nielsen's 10 usability heuristics with modern 2025+ interpretations, evaluation methodology, and severity rating framework. Foundation for all UX evaluation work.

Unified SEO skill with two modes: Article/Page Optimization and Full Website Audit. MODE 1 — ARTICLE/PAGE SEO OPTIMIZATION: Use this mode whenever the user shares an article, blog post, draft, page content, or a URL to a single page and mentions SEO, search optimization, ranking, or Google — or even when they just say "check this article", "optimize this", or "improve this" without explicitly saying SEO. Also trigger when the user pastes or attaches article text and asks you to review, improve, rewrite, or polish it for publishing. Trigger phrases include: "optimize this article", "SEO optimize this", "improve this for SEO", "optimize this page", "rewrite for SEO", "check this article", "improve this content", "make this rank", "SEO this". When in doubt, lean toward activating this mode — it's better to offer SEO optimization than to miss the opportunity. MODE 2 — FULL WEBSITE AUDIT: Use this mode whenever the user asks for a full website SEO audit, site-wide SEO strategy, multi-page SEO analysis, SEO health check, website audit, site audit, domain SEO review, or wants to understand their overall website SEO performance. Trigger phrases include: "audit my site", "SEO strategy for my website", "check my website SEO", "full SEO audit", "multi-page SEO", "site SEO", "website SEO review", "SEO health check", "site audit", "domain audit", or when the user provides a root URL and asks for SEO help. This mode crawls multiple pages across a website and produces a comprehensive HTML report with site-wide analysis, cross-page patterns, architecture review, and prioritized strategy.