adversarial-case-builder

Construct an adversarial case memo from evidence ledger and auditor findings.

123|7|Updated Jun 26, 2026
One-click install
npx skills add https://github.com/wanshuiyin/Anti-Autoresearch --skill adversarial-case-builder
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: adversarial-case-builder
Source: https://github.com/wanshuiyin/Anti-Autoresearch/tree/main/skills/adversarial-case-builder
Command: npx skills add https://github.com/wanshuiyin/Anti-Autoresearch --skill adversarial-case-builder

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill constructs and validates a compelling, evidence-bound case against a research paper, helping to ensure its accuracy and reliability.

Core Features & Use Cases

  • Adversarial Case Construction: Synthesizes the strongest evidence-bound case for rejecting a paper, using only evidence from the evidence ledger and other auditors' findings.
  • Defense Decomposition: Allows a separate reviewer to dissect the attack and rule each point against the anchored evidence.
  • Memo Generation: Produces a span-anchored, reviewer-ready memo that is used by the adjudicator without adding any verdict weight.
  • Use Case: Imagine you are an area chair reviewing a paper. Use this Skill to generate a detailed memo outlining the strongest evidence against the paper, and then defend it point-by-point against the anchored evidence.

Quick Start

Run the adversarial-case-builder skill on the paper directory and claims.json file to generate an adversarial case memo.

Frequently Asked Questions about adversarial-case-builder

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

FAQPage Schema
How do I build an evidence-bound adversarial case against a research paper?

To build an evidence-bound adversarial case, the Skill synthesizes findings from an evidence ledger and other auditors, constructing the strongest point-by-point argument for rejecting a paper. It outputs a span-anchored memo for adjudicator review.

Can I generate a reviewer-ready memo for the paper review process?

Yes, you can generate a reviewer-ready memo by running the Skill on the paper directory and claims.json file. It produces a span-anchored document detailing the synthesized adversarial evidence without adding extra verdict weight.

What is the defense decomposition process in an adversarial paper review?

Defense decomposition allows a separate reviewer to dissect the constructed attack against the paper. The reviewer rules each point individually by comparing it directly against the anchored evidence from the evidence ledger.

How does an area chair use an adversarial case during paper review?

An area chair uses this to generate a detailed memo outlining the strongest evidence against a paper. The chair can then defend this adversarial case point-by-point against the anchored evidence to validate findings.

Do I need an evidence ledger to construct an adversarial case against a paper?

Yes, the adversarial case construction relies strictly on an evidence ledger and other auditors' findings. It synthesizes only this anchored evidence to build the rejection argument and generates the final memo.

What are the limitations of using an adversarial case memo in the review process?

The generated memo serves strictly for adjudicator review without adding any verdict weight. It is bound entirely by the evidence ledger and auditor findings, preventing the introduction of external claims into the review process.