探索我们完整的 Claude 技能集合,扩展 AI 代理的能力。
Optimizes LLM interactions through advanced prompting techniques like few-shot learning, chain-of-thought, and systematic template design.
Implements a multi-layered validation strategy to catch bugs at every system boundary and make structural failures impossible.
Generates granular, step-by-step implementation roadmaps with exact file paths and test-driven tasks for seamless execution.
Enforces a mandatory protocol for discovering, announcing, and executing specialized skills to ensure rigorous development standards.
Generates hyper-detailed, step-by-step implementation plans using TDD principles and exact file paths for seamless feature development.
Systematically traces bugs backward through the call stack to identify and fix the original trigger rather than just treating symptoms.
Generates granular, step-by-step implementation plans with exact file paths and TDD instructions for engineers with zero codebase context.
Eliminates flaky tests by replacing arbitrary timeouts with smart condition polling to ensure reliable and fast async operations.
Applies a rigorous four-phase scientific framework to identify root causes and implement reliable fixes for complex software bugs.
Orchestrates multiple concurrent Claude agents to investigate and resolve independent codebase failures simultaneously.
Transforms rough conceptual ideas into structured technical designs through Socratic questioning and comparative architectural exploration.
Creates and manages isolated Git worktrees with automated dependency setup and safety verification for cleaner feature development.
Prevents common software testing errors by enforcing best practices for mocking, production code isolation, and test-driven development.
Enforces a strict evidence-based protocol requiring empirical verification before any task is marked as complete or successful.
Applies Test-Driven Development principles to documentation and skill creation to ensure bulletproof AI agent compliance.
Executes multi-task implementation plans by dispatching specialized subagents for each step with integrated code reviews.
Bridges Claude Code CLI with the Agent SDK to automate long-running development tasks and multi-feature implementations.
Analyzes code quality and manages SonarQube issues, quality gates, and security hotspots directly within your AI coding workflow.
Designs and implements robust, multi-stage CI/CD pipelines with integrated security checks, approval gates, and advanced deployment strategies.
Decomposes complex features into atomic user stories optimized for autonomous AI coding and single-iteration completion.
Manages fast, reproducible scientific Python environments by unifying the conda and PyPI ecosystems using the Rust-based Pixi tool.
Automates session status updates and progress tracking within the Claude Code environment for better task visibility.
Provides architectural patterns and conditional compilation strategies for sharing Swift code across the entire Apple ecosystem.
Manages GitLab issues, milestones, and project members directly through Claude using localized bash scripts.
Manages GitLab merge requests and code reviews directly through bash scripts within AI coding assistants.
Integrates SonarQube code quality and security analysis directly into Claude's workflow for automated issue tracking and remediation.
Conducts expert-level Android code reviews and posts automated inline comments directly to GitLab Merge Requests.
Converts Product Requirement Documents into structured Beads tasks for autonomous execution by Reeds.
Manages GitLab merge requests, reviews, and discussions directly within Claude Code using integrated bash scripts.
Manages GitLab issues and project workflows directly from the terminal using automated bash scripts.
Scroll for more results...