Anti-Autoresearch
NewDon't trust an autoresearch paper at face value. Reviewer-side integrity forensics — self-consistency + fabrication checks across 39 hack-patterns (7 families), deterministic verdict. Not an AI-text detector. The dual of ARIS.
Summary
This skill performs forensic integrity checks on academic papers, detecting self-contradictions and fabrication patterns across 39 hack-patterns in 7 families.
- It provides a deterministic verdict on trustworthiness, helping developers and reviewers avoid relying on flawed or fabricated research.
Install & Usage
mkdir -p .claude/skillsAdd the configuration to .claude/skills/anti-autoresearch.md
/anti-autoresearchUse Cases
Usage Examples
/anti-autoresearch check https://arxiv.org/abs/1234.56789
Run anti-autoresearch on the attached PDF and report any fabrication patterns found.
Analyze the paper 'Quantum Machine Learning for NLP' for self-consistency and hack-patterns.
Security Audits
Frequently Asked Questions
What is Anti-Autoresearch?
This skill performs forensic integrity checks on academic papers, detecting self-contradictions and fabrication patterns across 39 hack-patterns in 7 families. It provides a deterministic verdict on trustworthiness, helping developers and reviewers avoid relying on flawed or fabricated research.
How to install Anti-Autoresearch?
To install Anti-Autoresearch: create the skills directory (mkdir -p .claude/skills), then add the config to .claude/skills/anti-autoresearch.md. Finally, /anti-autoresearch in Claude Code.
What is Anti-Autoresearch best for?
Anti-Autoresearch is a other categorized under General. It is designed for: code-review, rust. Created by wanshuiyin.
What can I use Anti-Autoresearch for?
Anti-Autoresearch is useful for: Verify the internal consistency of a paper claiming novel results before citing it in your own work.; Check a literature review for fabricated references or data inconsistencies.; Audit a paper for common manipulation patterns like circular reasoning or false correlations.; Assess the integrity of a preprint before using it as a foundation for a code implementation.; Review a peer's submitted paper for signs of data fabrication or methodological flaws.; Evaluate a set of related papers for cross-paper consistency and potential coordinated fabrication..