# Peer Review Feedback

> Adversarial reading, soundness / novelty / clarity scoring, blameless feedback.

- Skill: `poly-gents/peer-review-feedback` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add poly-gents/peer-review-feedback`
- Raw SKILL.md: https://api.skillmd.com/api/skills/poly-gents/peer-review-feedback/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: poly-gents (https://skillmd.com/u/poly-gents)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/poly-gents/peer-review-feedback

---


# Peer review feedback

- Read once for **claim**, once for **proof**, once for **prose**.
- Score along three axes: **soundness** (is it correct?), **novelty** (does it add something?), **clarity** (can a non-author follow it?).
- For each issue, name the location (theorem N, line N), the problem, and a concrete suggested fix.
- Separate **fatal** (proof gap, false claim) from **major** (missing case, unclear statement) from **minor** (typo, notation).
- Tone is blameless and direct; assume the author is honest and tired.