Claude Skills

This skill provides a structured framework for developers to systematically analyze, refine, and optimize prompts for AI systems using evidence-based techniques. It helps eliminate anti-patterns and improve prompt structure, which is triggered by poor response quality or inconsistent outputs. The process includes A/B testing to validate effectiveness and produces an optimized prompt along with an analysis report.

This skill implements adaptive learning for AI agents using ReasoningBank to recognize patterns, optimize strategies, and enable continuous performance improvement. Use it when you need to enhance agent capabilities for repetitive tasks or strategy refinement. It outputs trained models, pattern libraries, and optimization recommendations with performance benchmarks.

This Claude Skill helps developers create new Claude Code Skills with proper YAML frontmatter and directory structure. It generates all required files including documentation and ensures skills follow best practices. Use this template generator when building reusable skills to maintain specification compliance and progressive disclosure.

This skill generates enterprise-grade PowerPoint presentations by enforcing structured workflows and design constraints. It produces accessible, professionally formatted decks with proper slide structure, notes, and WCAG compliance. Use it when you need automated generation of board decks, reports, or data-driven presentations.

This skill analyzes ambiguous user requests using cognitive science principles and probabilistic mapping to clarify intent. It automatically generates targeted questions to resolve uncertainties and produces actionable task definitions. Use it when handling complex, vague, or multi-part user instructions that require disambiguation.

This skill performs comprehensive codebase audits to detect placeholder code, mock data, and incomplete implementations like TODO markers or stub functions. It systematically identifies these "theater" instances, reads their full context, and completes them with production-quality code. Use it when you need to find and replace all non-production placeholders before deployment.

Use this Claude Skill for comprehensive quality verification during development cycles. It performs multi-dimensional testing including static analysis, dynamic testing, and integration validation with automated certification gates. The skill provides evidence-based quality metrics and creates audit trails to ensure code meets production standards.

This skill validates that generated code actually works through sandbox testing and execution verification. Use it after code generation to systematically identify and fix bugs while maintaining existing functionality. It creates isolated test environments, executes code with realistic inputs, and applies best practices for reliable fixes.

This skill orchestrates a multi-agent swarm to conduct comprehensive pull request reviews. It uses specialized agents to analyze code for security, performance, style, tests, and documentation in parallel. Use it for detailed, evidence-based feedback with auto-fix suggestions and a final merge readiness assessment.

This skill detects "theater code" - implementations that appear functional but fail during execution. It scans for suspicious patterns and validates actual functionality to prevent fake code from reaching production. Use it before merging AI-generated code or when code reviews reveal incomplete implementations.

This Claude Skill performs automated code style audits with auto-fix capabilities to enforce coding conventions across your entire codebase. It identifies style violations, validates formatting and naming conventions, and applies safe automated corrections. Use it for comprehensive style enforcement and maintaining consistent code quality throughout development.

Swarm Orchestration enables developers to coordinate multi-agent AI systems for parallel task execution using dynamic topologies like mesh and hierarchical structures. It provides intelligent coordination with automatic task distribution, load balancing, and fault tolerance. Use this skill when scaling beyond single agents, implementing complex workflows, or building distributed AI systems.

This skill provides advanced swarm orchestration patterns for managing complex distributed workflows in research, development, and testing scenarios. It enables developers to coordinate multiple specialized agents using both MCP tools and CLI commands for parallel task execution. Use it when you need to implement sophisticated multi-agent coordination across distributed systems.

The `style-audit` skill automatically rewrites functionally correct code to meet CI/CD style rules and quality guidelines, ensuring it is maintainable and production-ready. It enforces linting rules, naming conventions, and improves code organization through refactoring while preserving all verified behavior. Use this skill after functionality validation to finalize code quality before deployment.

The stream-chain skill enables multi-agent pipelines and sequential workflows by chaining JSON streams between processing steps. It provides both custom chains for flexible prompt sequences and predefined pipelines for common tasks, allowing each agent to receive the previous step's complete output. Use it for complex data transformations and coordinated multi-agent processing in Claude Code workflows.

The SPARC methodology provides a systematic development framework with 17 specialized modes for comprehensive software development from specification to completion. It integrates multi-agent orchestration to handle complex development workflows including architecture design, testing, and deployment. Use this skill when you need structured guidance throughout the entire development lifecycle with automated agent coordination.

This Claude Skill orchestrates a complete 10-week product launch by coordinating 15+ specialized agents across research, development, marketing, sales, and operations. It uses both sequential and parallel task orchestration to manage the complex workflow from planning through post-launch monitoring. Use this skill to automate and streamline multi-departmental product launch coordination.

This skill orchestrates a multi-agent code review process with specialized reviewers for quality, security, performance, and documentation. It executes a structured 4-hour workflow, starting with automated checks and progressing through specialized and integration reviews. Use it for comprehensive, automated code analysis when you need a thorough review across multiple technical domains.

This skill orchestrates a complete REST API development workflow using multi-agent coordination across backend, database, testing, and DevOps. It implements a structured 2-week timeline with a Test-Driven Development approach. Use it for end-to-end API projects requiring coordinated planning, development, testing, and deployment.

This skill provides an intelligent bug-fixing workflow that systematically identifies root causes using deep analysis and multi-model reasoning. It then generates fixes through Codex auto-fix and validates them with comprehensive testing and regression analysis. Use this skill for methodical debugging that combines automated fixes with thorough validation.

The slash-command-encoder creates fast, scriptable `/command` interfaces for accessing micro-skills, cascades, and agents. It automatically discovers skills and provides intelligent routing, parameter validation, and command chaining. Use this skill when you need efficient, unambiguous access for repeated operations over conversational interfaces.

skill-forge is an advanced skill creation system for Claude Code that transforms skill development from template filling into strategic design. It combines deep intent analysis, evidence-based prompting, and systematic engineering to build well-structured, sophisticated skills. Use it when creating or refining skills to ensure they follow best practices and incorporate advanced prompt techniques.

This skill creates Claude Code skills that spawn specialized agents optimized with evidence-based prompting techniques. Use it when you need reusable skills that leverage domain-specific agents for consistent, high-quality performance. It ensures each created skill communicates effectively with the parent Claude Code instance using best practices.

This Claude Skill helps developers create custom Claude Code Skills with proper YAML frontmatter and directory structure. It generates production-ready skill templates following Claude's specification for autonomous discovery across all surfaces. Use it when building custom workflows or when you need to understand the Claude Skills architecture.