Search Skills

Verifies codebase structure, file locations, and existing patterns to prevent design hallucinations during feature planning.

Assists users in determining the optimal generic agent type for specific AI task requirements.

Gathers current, well-sourced internet information to inform technical decisions, including API documentation, library patterns, and technology validation.

Creates and maintains CLAUDE.md files for project documentation, capturing architectural intent, contracts, and mandatory freshness dates.

Enforces mandatory workflows: Read tool pre-announcement, brainstorming before coding, TodoWrite checklists for task management.

Guides developers through final branch integration decisions including merge, pull request creation, or cleanup after successful implementation and testing.

Orchestrates initial design phases through context gathering, requirement clarification, brainstorming, and documentation to produce implementation-ready design artifacts.

Creates comprehensive implementation plans for engineers, including exact file paths, code examples, and verification steps for new codebase setup with minimal domain knowledge.

Automates code review requests by dispatching a reviewer subagent, handling retries, and managing the review-fix loop until all issues are resolved.

Enforces clarity, conciseness, and authenticity in technical documentation, guides, and API references for developer audiences while avoiding AI-generated patterns.

Writes validated design plans to documentation directory with structured format and discrete implementation phases after brainstorming.

Creates specialized subagents for Claude Code plugins and Task tool, covering prompt structure, tool selection, and agent testing for AI delegation.

Guides writing React components with TypeScript, hooks, and component composition while avoiding common pitfalls.

Enforces Functional Core, Imperative Shell pattern separation of pure logic from side effects with mandatory file classification during code refactoring.

Analyzes development changes to update stale project documentation files (CLAUDE.md, AGENTS.md) by identifying affected contracts and coordinating updates.

Creates isolated Git worktrees for feature development with workspace separation, smart directory selection, and safety verification.

Generates property-based tests to validate serialization, validation, normalization, and pure functions in code.

Empowers effective code writing and refactoring across multiple programming languages and runtimes with context-aware sub-skills.

Provides structured setup for Claude Code plugins including file organization, manifest format, and component definitions for commands, agents, skills, hooks, and MCP servers.

Provides a structured four-phase debugging framework (root cause investigation, pattern analysis, hypothesis testing, implementation) to understand issues before proposing fixes.

Orchestrates software implementation start: creates branches, plans details, and hands off with full context for execution.

Automates implementation plan execution by dispatching subagents for independent tasks, reviewing per phase, and loading phases just-in-time to minimize context usage.

Guides refinement of initial concepts into actionable designs using structured Socratic questioning and validation techniques.

Applies test-driven development to documentation, using subagents for iterative validation before finalizing skill documentation.