From claude-scholar
Runs pre-submission checks on LaTeX papers: verifies references, performs LaTeX cleanup, tests builds, scans figure formats, reviews front matter. Delivers unified report before journal submission or sharing drafts.
npx claudepluginhub yy/claude-scholar --plugin claude-scholarThis skill uses the workspace's default tool permissions.
Run a comprehensive set of checks on a LaTeX paper before submission.
Verifies citation references in LaTeX papers against academic databases using bibsleuth, flags unverified entries, and suggests missing DOIs/URLs. Use before submission or to check refs.
Validates arXiv paper submissions for compliance with TeX/LaTeX source, PDF, figures, metadata, bibliography, file organization, and error requirements. Produces pass/fail report with specific fixes.
Audits academic papers pre-submission using parallel agents for content, numerical consistency, references/DOIs, writing quality, figures/formatting, and replication archives.
Share bugs, ideas, or general feedback.
Run a comprehensive set of checks on a LaTeX paper before submission.
The skill runs these checks, ideally in parallel via subagents:
Verify all cited papers exist in academic databases. Flag unverified entries, suggest missing DOIs.
Run the full latex-cleanup checklist: common issues, style consistency, draft artifacts, cross-references, typography, and missing figures.
If a Makefile exists, run make and check for:
LaTeX Warning: Reference ... undefined)LaTeX Warning: Citation ... undefined)Scan for all figures included in the paper (\includegraphics, \input for pgf/tikz). For each figure file:
Read the paper and check:
\dataavailability, \codeavailability, or sections titled "Data Availability", "Code Availability", "Data and Code Availability", etc.Same search logic as check-refs:
paper/current/main.texpaper/main.texmain.tex**/*.tex with \begin{document}Launch subagents for independent checks:
.tex filemake in the paper directory if Makefile exists\includegraphics paths and flag bitmap formats (PNG, JPG, TIFF, BMP)Organize findings by severity:
Blockers (must fix):
Warnings (should fix):
draft mode still enabledFor user to verify (cannot be auto-checked):
Suggestions (nice to fix):
.bibFor each category, offer to fix what can be automated:
.bib patches (DOIs, URLs)After presenting the report and addressing formatting/technical issues, offer:
"Want me to run /critique-manuscript to check for content-level weaknesses reviewers might flag?"
This is a deeper, separate pass that evaluates the paper's substance (literature, methodology, claims, etc.) rather than its formatting and technical correctness.