Paper Reader Skill

Read an AI/CS academic paper and produce a rigorous Chinese-language analysis report at senior PhD researcher level. Output both a Markdown (.md) file and an elegantly typeset LaTeX (.tex) file that compiles with XeLaTeX.

Skill Files Location

This skill's supporting files are located relative to this SKILL.md:

• references/analysis-prompt.md — The full 10-section analysis prompt and writing guidelines
• references/latex-guide.md — LaTeX template usage guide, xeCJK pitfalls, and code examples
• assets/template.tex — The XeLaTeX template file
• scripts/setup-tex.sh — One-time TeX package installer
• scripts/extract_figures.py — Smart PDF figure extraction (uses PyMuPDF: caption detection + region rendering)
• scripts/extract_text.py — PDF text layer extraction (uses PyMuPDF: page-delimited plain text for grep-based verification)
• scripts/validate_figures.py — PDF-structure-based figure quality checks (default verifier; no extra deps)
• scripts/verify_extraction.py — OCR-based deep diagnostics (optional; requires tesseract)

IMPORTANT: At the start of every invocation, read references/analysis-prompt.md to load the full analysis structure and guidelines. This is mandatory — do not rely on memory of the prompt.