メニュー
Preprocessing CNS papers (PDF + HTML) to extract text, figures, and links. Outputs structured data for use by the paper-filter skill.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Acquire and verify data for CNS papers that passed paper-filter. Clones repositories, downloads and organizes datasets per evaluation setting, performs Level 4 deep verification (acquisition integrity, initial state, separability, consistency, per-setting completeness), creates an organized data environment. Updates filter_result.json with verification results, acquisition details, and task_info corrections.
Verify data components in filter_result.json after data-check. Checks Algorithm A boundary, evaluation setting validity, separability correctness, data directory integrity, and description consistency. Outputs a verification report for human review — does not modify filter_result.json or data directories.
Fix Dockerfile issues identified by batch_verify.sh. Reads verify_result.txt to diagnose import failures, version conflicts, and missing dependencies, then repairs the Dockerfile and updates packages.json. May also fix scripts (Script-First Rule) when packages are unavailable.
Verify paper-filter results by checking pass/reject judgment correctness, validating task_info field accuracy, and detecting internal inconsistencies. Outputs actionable corrections to filter_result.json.
Filter CNS papers for suitability in extracting machine learning tasks. Uses a three-level cascade funnel (Task Nature → Evaluation System → Data Completeness) with progressive filtering; stops immediately upon rejection. Input is preprocessed paper data; output is the filtering decision result.
Build a structured task package from a paper that has passed paper-preprocess, paper-filter, and data-check. Reorganizes data, generates problem description, evaluator, metadata, and environment Dockerfile.v3.
| name | paper-preprocess |
| description | Preprocessing CNS papers (PDF + HTML) to extract text, figures, and links. Outputs structured data for use by the paper-filter skill. |
| context | fork |
| agent | general-purpose |
| allowed-tools | Read, Bash(python *) |
Preprocess CNS papers using HTML (text/links) and PDF (figures) to extract structured data.
Before invoking this skill, provide:
*.pdf and *.html files for the same paper (Nature/Springer structured HTML). The folder name is used as paper_id.Use scripts/extract_text.py to extract full text from the HTML:
Use scripts/extract_figures.py to extract figures from the PDF:
Use scripts/extract_links.py to extract links from the HTML:
cd skills/paper-preprocess/scripts
python3 preprocess.py <input_dir> <output_dir>
# Example:
python3 preprocess.py ./s42256-019-0037-0/ ./s42256-019-0037-0/preprocessed
{output_dir}/
text.md # Full paper text (Markdown format)
figures/ # Figures directory
Fig_1_page2.png
Fig_2_page3.png
...
tables/ # Tables directory
Table_1_page3.png
...
links.json # Extracted links with section classification
{
"links": [
{
"url": "https://github.com/example/repo",
"type": "github | zenodo | codeocean | huggingface | google_drive | dropbox | other",
"context": "Surrounding paragraph text (up to 400 chars)",
"section": "data_availability | code_availability | supplementary_information | other"
}
],
"data_availability": ["https://..."],
"code_availability": ["https://github.com/..."],
"paper_id": "s42256-019-0037-0",
"extraction_timestamp": "2026-01-01T00:00:00"
}
Install Python dependencies before running:
pip install pymupdf beautifulsoup4