adversarial-review

Identify hidden design flaws and edge cases in code submissions with evidence-based reports.

8|1|Updated Mar 22, 2026

Search Tags:#edge-cases #quality-assurance #code-review #security-analysis #adversarial-review #independent-review

Ready to run

npx skills add https://github.com/zxdxjtu/harness --skill adversarial-review-zxdxjtu

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: adversarial-review
Source: https://github.com/zxdxjtu/harness/tree/main/skills/adversarial-review
Command: npx skills add https://github.com/zxdxjtu/harness --skill adversarial-review-zxdxjtu

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Independent code reviewers often miss subtle edge cases and exploitable design choices. This Skill enforces rigorous, decentralized review by an autonomous adversarial reviewer that isolates evaluation from optimistic outputs, ensures objective findings, and drives concrete fixes.

Core Features & Use Cases

Independent review: identify hidden defects, boundary conditions, and potential risks without developer bias.
Evidence-based findings: reports with code evidence and actionable remediation steps.
Use Case: apply to feature changes, architecture reviews, and pull requests to continuously raise quality and robustness.

Quick Start

Provide a feature-id to the adversarial-review Agent to start an independent code quality assessment.

Frequently Asked Questions about adversarial-review

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

FAQPage Schema

How do I find hidden design flaws and edge cases in a pull request?▼

Adversarial code review isolates evaluation from optimistic developer outputs to identify hidden design flaws and edge cases in feature changes and pull requests. It provides objective, evidence-based findings with concrete remediation steps.

What is an independent code review and how does it expose exploitable design choices?▼

An independent code review is a decentralized assessment that isolates evaluation from developer bias to expose exploitable design choices and boundary conditions. It enforces rigorous analysis to continuously raise code robustness.

How do I generate a structured defect report with severity levels for code submissions?▼

To generate a structured defect report with severity levels for code submissions, apply an autonomous adversarial reviewer to your repository changes. It produces evidence-based findings with concrete remediation steps.

Can I use adversarial review for architectural reviews across multiple repositories?▼

Yes, adversarial review applies to architectural reviews across repositories. It identifies potential risks and hidden defects in large-scale feature work by ensuring objective, evidence-based findings.

What is the best way to ensure objective findings during a security analysis of code changes?▼

The best way to ensure objective findings during security analysis is to apply an autonomous adversarial reviewer that isolates evaluation from optimistic outputs. This enforces rigorous, decentralized review without developer bias.

How do I start an independent code quality assessment for a new feature?▼

To start an independent code quality assessment for a new feature, provide a feature-id to the adversarial-review agent. It will autonomously assess code quality and expose hidden defects and boundary conditions.