LLM & Agents

Guides subagent coordination through implementation workflows. Use when orchestrating multiple agents, managing workflow phases, or determining autonomous execution mode. Defines scale determination, document requirements, and stop points.

Frontend testing rules with Vitest, React Testing Library, and MSW. Includes coverage requirements, test design principles, and quality criteria. Use when writing frontend tests or reviewing test quality.

Metacognition skill for AI agents. Use when starting work, feeling stuck, output feels off, or before complex tasks. Teaches how to think about thinking.

Testing pyramid and test writing standards for UT/IT/ST/E2E. Use when: writing tests, discussing test coverage, test strategy, or test naming. Keywords: test, unit, integration, e2e, coverage, mock, 測試, 單元, 整合, 端對端.

Guide requirement writing, user story creation, and feature specification. Use when: writing requirements, user stories, issues, feature planning. Keywords: requirement, user story, issue, feature, specification, 需求, 功能規劃, 規格.

Guide OAK constitution maintenance with amendment workflows, validation frameworks, semantic versioning, and agent instruction synchronization.

Systematic code review checklist and pre-commit quality gates for PRs. Use when: reviewing pull requests, checking code quality, before committing code. Keywords: review, PR, pull request, checklist, quality, commit, 審查, 檢查, 簽入.

Multi-agent coordination with Agent Mail MCP. Use when registering agents, reserving files, sending messages, coordinating with other agents, or when the user mentions "agent mail", "coordination", "file reservation", or "multi-agent".

Run an evidence-seeking calibration roundtable to realign the plan with the North Star. Use when pausing between phases, when agents disagree, when reviewing work, when the user mentions "calibrate" or "realign", or when making decisions that affect the plan.

New agent startup checklist for Agent Mail and Beads. Use when starting a new agent session, when the user says "prime" or "startup", or when beginning work on a multi-agent project.

自动同步项目中的 Agents.md、claude.md 和 gemini.md 文件，保持内容一致性。支持自动监听和手动触发。

Resolve disagreements between agents or approaches using test-based adjudication. Use when agents disagree, when multiple valid approaches exist, when the user asks "which approach", or when making architectural decisions with tradeoffs.

AWS Bedrock AgentCore comprehensive expert for deploying and managing all AgentCore services. Use when working with Gateway, Runtime, Memory, Identity, or any AgentCore component. Covers MCP target deployment, credential management, schema optimization, runtime configuration, memory management, and identity services.

Skill for creating and editing Youtube thumbnails that are optimized for click-through rate. This skill should not be used directly, instead use the Thumbnail Designer subagent who can also invoke this skill. Use when the user asks to create a thumbnail from scratch or edit an existing thumbnail.

Generate a project-specific AGENTS.md from a user goal, then confirm before overwriting.

Build Retrieval-Augmented Generation (RAG) systems for AI applications with vector databases and semantic search. Use when implementing knowledge-grounded AI, building document Q&A systems, or integrating LLMs with external knowledge bases.

Model Context Protocol (MCP) server implementation patterns with LangChain4j. Use when building MCP servers to extend AI capabilities with custom tools, resources, and prompt templates.

Build declarative AI Services with LangChain4j using interface-based patterns, annotations, memory management, tools integration, and advanced application patterns. Use when implementing type-safe AI-powered features with minimal boilerplate code in Java applications.

Integration patterns for LangChain4j with Spring Boot. Auto-configuration, dependency injection, and Spring ecosystem integration. Use when embedding LangChain4j into Spring Boot applications.

This skill should be used when creating, optimizing, or implementing advanced prompt patterns including few-shot learning, chain-of-thought reasoning, prompt optimization workflows, template systems, and system prompt design. It provides comprehensive frameworks for building production-ready prompts with measurable performance improvements.