Skill UI

Get a multi-round adversarial critical review of your research ideas, papers, or experimental results from an external LLM reviewer backend (Codex or manual). Compiles a briefing, runs ultra-reasoning review rounds, and returns a verdict with logical gaps, missing experiments, and narrative weaknesses.

Perform a two-thread adversarial review: a fresh reviewer writes a strong 200-word rejection memo, then another fresh reviewer defends the paper point-by-point and classifies unresolved issues. Use it before theory paper submission or rebuttal preparation to surface the most damaging possible reviewer attack.