When is the best time to trigger the /amplihack:debate command?

Use it for high-impact decisions that are difficult to reverse, such as selecting a primary database, choosing between REST and GraphQL, or deciding on a deployment strategy.

Does this skill produce permanent documentation?

Yes, it automatically generates a detailed Markdown decision record in your 'decisions/' directory, including the full debate transcript, rationale, and dissenting opinions.

Can I customize which perspectives are involved in a debate?

Yes, the skill supports a range of profiles including Security, Performance, Simplicity, Maintainability, UX, Scalability, and Cost, which can be configured based on the complexity of the decision.

How does the debate workflow prevent model bias?

The skill spawns independent Claude subprocesses for each perspective (e.g., Security, Performance) without initial context sharing, forcing each 'agent' to form an objective position based solely on their assigned profile.

Architectural Debate Workflow

Name: Architectural Debate Workflow
Author: rysweet

byrysweet

•

Collaboration Tools

Facilitates structured multi-perspective debates to resolve complex architectural decisions and technical trade-offs using specialized AI agents.

The Debate Workflow skill implements a rigorous decision-making framework within Claude Code, enabling developers to simulate various professional perspectives such as security, performance, and maintainability. By spawning independent agent subprocesses to challenge and refine technical proposals through multiple rounds, it helps teams move past stalemates on controversial changes or framework selections. This workflow ensures that high-impact decisions are evidence-based, thoroughly documented via automated Decision Records (ADRs), and transparent for all stakeholders.

Key Features

01Independent subprocess spawning to ensure objective, non-biased evaluation

02Customizable convergence criteria including consensus, majority, or synthesis

0317 GitHub stars

04Multi-perspective simulation involving Security, Performance, and Simplicity personas

05Automated Decision Record (ADR) generation for project history

06Iterative debate cycles including initial positions, challenges, and responses

Use Cases

01Evaluating major framework selections or system design overhauls

02Resolving technical stalemates on controversial refactoring proposals

03Assessing complex infrastructure trade-offs like database migrations or scaling strategies

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add rysweet/amplihack debate-workflow

For use in Claude.ai and ChatGPT

Download Skill