What problem does it solve?
This Skill helps authors get rigorous, venue-style feedback on their own AI/ML paper draft before submission, reducing the risk of avoidable clarity gaps, methodological issues, and presentation problems.
Core Features & Use Cases
- Project auto-resolution + LaTeX ingestion: Resolves a local Overleaf project from a provided name, recursively ingests LaTeX sources and pulls structured elements (sections, figures, tables, equations, citations).
- Reviewer-style self-critique pipeline: Performs a multi-phase review that starts from a steel-manned best-interpretation summary and then interrogates sections with specificity.
- Mechanical preflight to protect reliability: Runs compilation, page-limit checks, build/log scans, placeholder red-flag greps, and citation hygiene checks before producing the review report.
- Produces an actionable report: Writes a reviewer-style markdown report to an underscore-prefixed
_reviews/ directory with classified issues.
Quick Start
Ask for a pre-submission review of the Overleaf draft by providing the project name, e.g., “review my paper draft for submission” and include the name of the folder under ~/Documents/docs/project/.