Welcome to llm-workflow#

The llm-workflow repository provides a collection of generated platform skills designed to enhance LLM-assisted coding workflows. These skills extend Claude Code’s capabilities with specialized knowledge, workflows, and tools for common development tasks.

What are Skills?#

Skills are modular, self-contained packages that extend supported LLM coding platforms’ capabilities. Think of them as “onboarding guides” for specific domains or tasks—they transform Claude from a general-purpose agent into a specialized agent equipped with procedural knowledge.

Skills can provide:

  • Specialized workflows - Multi-step procedures for specific domains

  • Tool integrations - Instructions for working with specific file formats

  • Domain expertise - Project-specific knowledge and business logic

  • Bundled resources - Scripts, references, and assets for complex tasks

Available Skills#

Currently, the following skills are available:

  • Code Review Response - Process manual code review comments marked with ISSUE: prefix into structured reports

  • Complexity - Cyclomatic complexity metrics (radon/gocyclo) and ranked refactor targets

  • Council - Multi-judge validation council using Claude Code agent teams for adversarial review from independent context windows

  • Debate - Orchestrate structured multi-round debates between two LLM personas via HydraMCP, producing a transcript with synthesis

  • Pre-Mortem - Adversarial pre-mortem failure analysis with 4 failure-specialist judges and conservative unanimous verdict threshold

  • Team Review Implementation - Team-based implementation-vs-spec review with 3 specialist judges verifying code delivers what the spec describes

  • Thermo-Nuclear Code Quality Review - Strict maintainability review for code judo, giant files, and spaghetti-condition growth

Getting Started#

See Installation to set up the plugin, then Usage for how to invoke skills in your workflow.

Contents#

Indices and tables#