Claude Skills

Initialize a threat modeling project by analyzing architecture documentation. Creates threat model structure with asset inventory, data flows, trust boundaries, and attack surface mapping. Use when starting new threat modeling work, setting up threat model for a project, or creating initial security assessment.

Review documentation changes for compliance with the Metabase writing style guide. Use when reviewing pull requests, files, or diffs containing documentation markdown files.

Master advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability in production. Use when optimizing prompts, improving LLM outputs, or designing production prompt templates.

Create well-formatted git commits with conventional commits format

Display current token usage and budget status with recommendations

Display all available Potenlab commands with descriptions

Configure project with MCPs, rules, and agents in one command

Refactor code with clear before/after explanation. Verifies changes with tests before completion.

Creates a GitHub PR for current work. Handles branch creation, committing, pushing, and PR creation.

Refactor high-complexity React components in Dify frontend. Use when `pnpm analyze-component --json` shows complexity > 50 or lineCount > 300, when the user asks for code splitting, hook extraction, or complexity reduction, or when `pnpm analyze-component` warns to refactor before testing; avoid for simple/well-structured components, third-party wrappers, or when the user explicitly wants testing without refactoring.

Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.

**URL页面集成测试技能** 本技能提供由 `agent-browser` CLI 驱动的原子级浏览器自动化工具。它作为AI Agent的“眼睛”和“双手”，通过快照 + ref 工作流进行逻辑驱动的网页功能集成测试，具有高可靠性和效率。 **核心工作流简介**： 1. 打开目标URL 2. 执行快照以获取带有唯一ref（@e1、@e2 等）的交互元素 3. 使用ref执行动作（填写、点击等） 4. 页面发生变化后重新快照（之前的ref会失效） 5. 根据结果生成Markdown报告（参考 `references/REPORT_GUIDE.md`）

Refactor high-complexity React components in Dify frontend. Use when `pnpm analyze-component --json` shows complexity > 50 or lineCount > 300, when the user asks for code splitting, hook extraction, or complexity reduction, or when `pnpm analyze-component` warns to refactor before testing; avoid for simple/well-structured components, third-party wrappers, or when the user explicitly wants testing without refactoring.

Creates new Claude Code skills with best practices. Use when the user wants to create a skill, add a slash command, or automate a workflow.

Required folder structure and naming conventions for theme implementations.

Design and deploy Model Context Protocol servers with deep knowledge of the JSON-RPC 2.0 spec, transport mechanisms, and protocol lifecycle. Covers FastMCP framework patterns, production hardening, and diagnostic techniques for connectivity failures. Engage when architecting MCP integrations, building custom servers, or debugging protocol-level issues.

Design new agent skills using the 10-step Skill Design Loop (SDL). Use when creating reusable, portable agent capabilities with progressive disclosure, VOI-driven uncertainty management, and security hardening.

Export fiction manuscripts to professional book formats (PDF, ePub) with research-informed styling profiles. Includes profiles for manuscript drafts, beta readers, and publication-ready output.