skills.rest
leonletto
leonlettoCOMMUNITY
Explore all skills by this author →
Website𝕏@leonlettoGitHub

adversarial-critique

Orchestrate adversarial critiques to resolve design forks with a defended choice.

28|5|Updated Feb 3, 2026
Search Tags:#multi-agent#debate#critique#adversarial#audit-trail#design-decision
One-click install
npx skills add https://github.com/leonletto/thrum --skill adversarial-critique
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: adversarial-critique
Source: https://github.com/leonletto/thrum/tree/main/opencode-plugin/assets/skills/adversarial-critique
Command: npx skills add https://github.com/leonletto/thrum --skill adversarial-critique

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Use when an implementer is blocked mid-work by a design flaw, contradiction, or unforeseen fork with 2-3 ways to resolve it - produces a defended pick that unblocks the work and leaves an audit trail so the decision can be revisited if it later proves wrong

Core Features & Use Cases

Stress-test a mid-implementation design fork by dispatching sub-agents that argue opposing options across three structured rounds. Produces a defended pick with invariants and acceptance criteria — not a "both approaches work" summary. The artifact left on disk is an audit trail: if the decision proves wrong later, future readers can see exactly why it was made and what was considered.

Quick Start

Run three parallel agent evaluations to generate a complete debate transcript and save it to dev-docs/brainstorms/YYYY-MM-DD-adversarial-critique-debate.md

Frequently Asked Questions about adversarial-critique

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I resolve a mid-implementation design fork with multiple viable options?▼

To resolve a mid-implementation design fork, you can orchestrate an adversarial critique where sub-agents argue opposing options across structured debate rounds, yielding a defended choice with an auditable trail.

What is an adversarial critique and how does it help with design decisions?▼

An adversarial critique is a structured debate mechanism that stress-tests design decisions by dispatching sub-agents to argue opposing options. It produces a defended pick with invariants and acceptance criteria instead of a generic summary.

How do I document design decisions so they can be revisited if proven wrong later?▼

You document design decisions by generating a complete debate transcript that outlines why the choice was made and what was considered, saving the audit trail directly to dev-docs/brainstorms/YYYY-MM-DD-adversarial-critique-debate.md.

When should I use a multi-agent debate for resolving design blockers?▼

Use a multi-agent debate when you are blocked mid-work by a design flaw, contradiction, or unforeseen fork with 2-3 viable options that require explicit rationale and structured argumentation to unblock implementation.

Does adversarial critique work for design forks with more than three options?▼

Adversarial critique is designed specifically for mid-implementation forks with 2-3 viable options. It applies structured debate across three rounds to ensure explicit rationale rather than handling an unlimited number of alternatives.

Related Skills

Explore More Design & Creative Skills→
rafaelcallejarafaelcalleja
2

simplification-cascades

Identify unifying insights that eliminate multiple components in complex systems.

Community
Basic
Cogni-DAOCogni-DAO

spec

Create and manage technical specifications with markdown and lifecycle states.

Official
Intermediate
starlink-awakenstarlink-awaken

Council

Coordinate multi-agent debates with round-by-round transcripts and final synthesis.

Community
Advanced
max4cmax4c
8

design-interface

Generate 3-5 distinct UX design explorations in parallel for a feature.

Community
Advanced