Use when code changes need multi-pass automated review before merge. Use when preparing branch for PR. Use when thorough code quality check needed.

Learn ADK concepts and understand the retail agent pipeline. Use when asking how the agent works, what ADK is, wanting to understand the architecture, state flow, or individual agent purposes.

Create OpenAI Agents SDK applications in TypeScript/JavaScript. Use when building AI agents, multi-agent systems, voice agents, or any agentic workflow with the OpenAI Agents SDK. Covers agents, tools, handoffs, guardrails, streaming, MCP, and realtime voice agents.

Build elegant frontend UIs following Microsoft Foundry's NextGen Design System using Vite + React + pnpm. Use when creating dashboards, agent builders, data grids, entity management interfaces, or any application matching Foundry's refined dark-themed aesthetic. Triggers on requests for Foundry-style UI, NextGen design system, Microsoft Foundry interfaces, or enterprise admin dashboards with data tables, detail panels, and charts.

This skill should be used when working with Rust code, reviewing Rust code, managing Rust dependencies, creating Rust projects, or fixing Rust compilation errors. It provides strict coding standards (especially FAIL FAST error handling), workspace architecture guidance, dependency management automation, and common Rust patterns.

Multi-agent task orchestration for Claude Code. When given multiple tasks to execute, use this skill to spawn parallel worker agents through Sibyl's MetaOrchestrator instead of Claude Code's built-in Task tool. This enables true multi-process execution with budget controls, quality gates, and coordination.

Architectural code analysis for Python design quality. Evaluates simplicity (Rich Hickey), functional core/imperative shell (Gary Bernhardt), and coupling (Constantine & Yourdon). Use for design review or architectural assessment of Python code.

````skill

Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).

Based on the Recursive Language Models (RLM) research by Zhang, Kraska, and Khattab (2025), this skill provides strategies for handling tasks that exceed comfortable context limits through programmatic decomposition and recursive self-invocation. Triggers on phrases like "analyze all files", "process this large document", "aggregate information from", "search across the codebase", or tasks involving 10+ files or 50k+ tokens.

Enforce JavaScript/ES6+ and TypeScript coding standards including camelCase variables, PascalCase classes, and kebab-case filenames.

Generate Go Bubble Tea TUI applications with chat-like interfaces. Use when the user asks to create terminal UIs, chat interfaces, interactive CLI, REPLs, or log viewers. Follow the specified architecture and implementation requirements for consistent results.

Expert code review for Python focused on correctness, maintainability, error handling, performance, and testability. Use after writing or modifying Python code, or when reviewing refactors and new features.

Analyzes recently modified code and creates pull requests with simplifications that improve clarity, consistency, and maintainability while preserving functionality

> **版本演進**：

Contribute to Agentpack itself (spec-driven changes, tests, releases).

This skill should be used when tracing ManifestWork resources through the Maestro system to find relationships between user-created work names, resource IDs, and applied manifests, or to debug manifest application issues across the management cluster and database.

Frontend UI/UX design and implementation for HTML/CSS/JS including semantic structure, responsive layout, accessibility compliance, and visual design direction. Use for building or reviewing web pages/components, fixing accessibility issues, improving styling/responsiveness, or making UI/UX decisions.

Generate an executive assessment report from GitHub Quick Review (ghqr) scan data. Produces an executive summary, a dedicated section per validated subject with all findings, and a prioritized 30/60/90-day remediation plan. Use when the user asks for a report, executive summary, best practices posture overview, or a remediation roadmap from ghqr scan results.

Overview of protected quality thresholds and quick reference for all quality tools. Use when you need to understand quality metrics, run comprehensive quality checks, or learn which specialized skill to use. For specific issues, use dedicated skills (deptrac-fixer for Deptrac, complexity-management for PHPInsights, testing-workflow for coverage).