Comprehensive manuscript review covering technical correctness, proof validity, security analysis, and presentation quality. Simulates a top-venue referee report.
Produce a thorough, constructive review of an academic manuscript in cryptography or mathematics — the kind of report a top-venue (CRYPTO, EUROCRYPT, CCS) referee would write.
Input: $ARGUMENTS — path to a paper (.tex or .pdf).
Read the full paper end-to-end. For large PDFs, read in chunks (5 pages at a time).
Evaluate across 6 dimensions (see below).
Generate 3-5 "referee objections" — the tough questions a program committee member would ask.
Save the review to review_[sanitized_name].md in the current directory.
# Review: [Paper Title]
**Date:** [YYYY-MM-DD]
## Summary Assessment
**Recommendation:** [Strong Accept / Accept / Borderline / Reject]
[2-3 paragraph summary]
## Strengths
1. [Strength]
## Major Concerns
### MC1: [Title]
- **Dimension:** [Correctness / Novelty / Security / Efficiency / Literature / Presentation]
- **Issue:** [Specific description]
- **Suggestion:** [How to address]
## Minor Concerns
### mc1: [Title]
- **Issue / Suggestion**
## Referee Objections
### RO1: [Question]
**Why it matters:** [Why this could be fatal]
**How to address:** [Suggested response]
## Ratings
| Dimension | Rating (1-5) |
|---|---|
| Technical Correctness | |
| Novelty | |
| Security Model | |
| Efficiency | |
| Literature | |
| Presentation | |
Adapted from pedrohcgs/claude-code-my-workflow (MIT license).