Managed Deep Agents

Structure software around the Dependency Rule: source code dependencies point inward from frameworks to use cases to entities. Use when the user mentions "architecture layers", "dependency rule", "ports and adapters (hexagonal)", "onion architecture", "screaming architecture", "where should business logic go", "decouple from the database", "swap the framework without a rewrite", or "keep business rules independent". Also trigger when deciding which layer code belongs in, isolating core logic from infrastructure, defining module boundaries, or debating whether the framework should call your code or the reverse. Covers component principles, boundaries, and SOLID. For code-level quality, see clean-code. For domain modeling, see domain-driven-design.

演讲铸造器（Outline-Faithful）。把 orgmode/markdown outline 1:1 铸成单文件离线 HTML；支持 black/red/yellow、浅色 hacker 与暗色 hacker-dark，自动处理标题封面、多行密度布局、表格、ASCII、LaTeX、自适应与翻页笔。USE WHEN 用户要求讲这个、present、做成演讲、slides、标语流、宣言体、slogan、manifesto、按 outline 美化。NOT FOR 内容提炼、改写或企业 PPT。

AI-powered codebase security scanner that reasons about code like a security researcher — tracing data flows, understanding component interactions, and catching vulnerabilities that pattern-matching tools miss. Use this skill when asked to scan code for security vulnerabilities, find bugs, check for SQL injection, XSS, command injection, exposed API keys, hardcoded secrets, insecure dependencies, access control issues, or any request like "is my code secure?", "review for security issues", "audit this codebase", or "check for vulnerabilities". Covers injection flaws, authentication and access control bugs, secrets exposure, weak cryptography, insecure dependencies, and business logic issues across JavaScript, TypeScript, Python, Java, PHP, Go, Ruby, and Rust.

Configure a PreToolUse hook to prevent AI agents from skipping git pre-commit hooks with --no-verify and other bypass flags. Use when setting up Claude Code projects that enforce commit quality gates.

Streaming backends and wire protocols for assistant-ui via the assistant-stream package. Use when building a custom (non-AI-SDK) streaming endpoint with createAssistantStreamResponse or createAssistantStreamController, emitting parts through appendText/appendReasoning/appendSource/appendFile/addToolCallPart and setResponse; choosing between the AI SDK UI-message format (toUIMessageStream + createUIMessageStreamResponse) and the native Assistant Transport format; encoding or decoding streams with DataStreamEncoder/DataStreamDecoder, AssistantTransportEncoder/AssistantTransportDecoder, PlainTextEncoder, or UIMessageStreamDecoder; or wiring streamed chunks into useLocalRuntime or useChatRuntime. Use specifically for debugging stream wire issues: text-delta, part-start, result events, text/event-stream content-type, SSE format, tool calls not rendering, or partial text not showing. For general non-stream debugging route to the relevant focused skill, not the parent overview.

Deploy, manage, inspect, and troubleshoot Vercel projects from the command line. Use for Vercel deployments, build failures, projects and teams, environment variables, domains and DNS, logs, metrics, Speed Insights, Core Web Vitals, request traces, usage, activity, alerts, firewall rules, cache, cron jobs, deploy hooks, Edge Config, feature flags, integrations, connectors, Blob storage, Container Registry (VCR), microfrontends, rolling releases, custom environments, Sandbox, agent/MCP setup, OAuth apps, preview access, local development, or `vercel api` fallback.