政府公文格式文档生成器。支持所有常见文档格式输入（.md .txt .docx .pdf .html .rtf），输出符合国家标准的.docx公文格式。使用场景：产业调研报告、政策分析报告、工作报告等需要严格按照政府公文格式排版的文档。

This skill appears to do what it says: convert local documents into a government-style .docx using python-docx and some XML patching. Before installing/using: (1) review and install required Python packages (python-docx, pdfminer.six, beautifulsoup4, striprtf or equivalents) in a controlled environment; (2) inspect the remainder of scripts/convert.py (the provided file was partially truncated in the review) to confirm there are no network calls, subprocess.exec/os.system calls, or attempts to read unrelated filesystem locations; (3) run conversions on non-sensitive test documents first to verify formatting and that content isn't unintentionally modified (the script performs aggressive punctuation/format cleanup and will remove bold/links and certain characters); (4) keep originals backed up; and (5) run in an isolated environment (virtualenv or container) if you have security concerns. If you want, I can scan the remainder of convert.py (the truncated portion) for unexpected behavior or network access.

Type: OpenClaw Skill Name: gongwen-writer Version: 1.4.0 The 'gongwen-writer' skill is a specialized document converter designed to produce Chinese government-standard .docx files from various input formats (Markdown, PDF, HTML, etc.). The core logic in `scripts/convert.py` uses standard libraries like `python-docx`, `pdfminer.six`, and `BeautifulSoup4` to parse text and apply strict formatting rules. A low-level ZIP manipulation in the `_patch_settings_even_odd` function is used to enable specific Word features (odd/even page footers) not supported by the high-level API, which is clearly documented as a technical workaround in `lessons-learned.md`. No malicious intent, data exfiltration, or suspicious execution patterns were detected.