探索我们完整的 Claude 技能集合,扩展 AI 代理的能力。
Eliminates common testing mistakes by enforcing strict TDD principles and preventing mock-related technical debt.
Architects and deploys specialized AI assistants within Claude Code to delegate complex tasks and maintain isolated execution contexts.
Decomposes complex features into granular, independently testable sub-tasks organized by implementation layer and development phase.
Performs ultra-critical, multi-pass analysis of code changes to identify runtime failures, architectural flaws, and environment risks before merging.
Streamlines the creation of new Claude skills that interface with Model Context Protocol (MCP) servers using optimized performance patterns and automated directory structures.
Guides the end-to-end development, structuring, and packaging of specialized knowledge and automated workflows for Claude Code.
Provides a standardized educational framework for building custom Claude Code skills using progressive disclosure patterns and modular directory structures.
Applies Test-Driven Development principles to create and verify robust, loophole-free documentation for Claude Code skills.
Tailors Claude's interaction style and personality across four distinct modes to match user mood, task complexity, or professional requirements.
Enables comprehensive observability for Claude Code by automatically tracing sessions, LLM calls, and tool usage to Judgeval.
Writes strategic, behavior-focused tests for core user flows and critical business paths to balance development velocity with reliable coverage.
Analyzes new requirements against existing specifications and tasks to prevent duplicate efforts and identify architectural overlaps.
Systematically traces bugs through the call stack to identify original triggers and implement long-term fixes.
Enforces code quality standards by running linting, type-checking, and test suites before tasks can be marked as completed.
Conducts multi-source research across codebases and the web to provide deep architectural analysis and complex debugging solutions.
Provides a comprehensive understanding of the Agent Runtime architecture, purpose, and implementation patterns for orchestrating isolated AI agent workloads.
Implements multi-tiered data validation strategies to prevent deep execution failures and make software bugs structurally impossible.
Configures and deploys robust agent runtime backends using Node.js, Modal sandboxes, and real-time WebSocket communication.
Deploys and orchestrates cloud-based AI agent swarms with event-driven workflow automation and intelligent coordination.
Orchestrates complex software development through a structured specification, design, and test-driven implementation framework using multi-agent coordination.
Enforces mandatory workflows for skill discovery, brainstorming, and task tracking to ensure consistent and high-quality AI output.
Automates the transformation of planning analysis into formal, structured technical specification documents.
Transforms Claude into a collaborative partner that prioritizes iterative development, frequent feedback, and shared decision-making over autonomous implementation.
Defines agent profiles, capabilities, and orchestration workflows to customize AI assistant behavior and tool integrations.
Implements complex development plans through controlled batch execution and mandatory review checkpoints.
Transforms rough ideas into structured technical designs through iterative collaborative dialogue and incremental validation.
Builds production-grade software using elite engineering protocols for scalability, security, and performance.
Integrates React applications with agent runtimes using specialized hooks and context providers for real-time communication and session management.
Transforms high-level specification documents into granular, dependency-mapped, and complexity-scored task lists for automated execution.
Orchestrates multi-phase investigations across codebases and documentation using specialized parallel agents to generate evidence-based insights.
Scroll for more results...