Academic Paper Analysis

Extract and analyze academic/scientific PDF papers.

When to use this skill

• •User asks to analyze or review a research paper
• •User wants methodology, findings, or conclusions extracted from a paper
• •User needs the references/bibliography section listed
• •User asks "what is this paper about?"

Steps

1. •

   Ensure pymupdf is installed:

   bash

   pip install pymupdf
   

2. •

   Extract text from the paper.

   bash

   python scripts/extract_text.py "INPUT_FILE_PATH"
   

   The script prints extracted text directly to stdout. You do NOT need to read a separate file — just use the shell output.

3. •

   Read the shell output (or use the Read tool if saved to file).

4. •

   Produce a structured analysis:

   code

   ## Paper Analysis
   
   **Title**: ...
   **Authors**: ...
   **Published**: journal/conference, year
   
   ### Abstract Summary
   Problem → Method → Result (1–2 sentences each)
   
   ### Research Question / Hypothesis
   What the authors set out to prove or investigate.
   
   ### Methodology
   - Study design (experimental, survey, simulation, etc.)
   - Data sources and sample size
   - Tools / frameworks used
   
   ### Key Findings
   - Finding 1 (with statistical significance if reported)
   - Finding 2
   
   ### Limitations
   Issues acknowledged by the authors or identified in review.
   
   ### References of Note
   Key citations relevant to understanding this work.
   

5. •

   (Optional) Extract only the references section:

   bash

   python scripts/extract_references.py "INPUT_FILE_PATH"
   

Edge cases

• •Multi-column layouts: pymupdf may interleave columns. Read carefully and reconstruct logical flow.
• •Supplementary materials: If the user mentions appendices, they may be in a separate PDF.
• •Non-English papers: pymupdf extracts Unicode text; analysis quality depends on the language model's capabilities.

Scripts

• •extract_text.py — Extract full paper text
• •extract_references.py — Extract References/Bibliography section