TDOC Docx

PassAudited by ClawScan on May 1, 2026.

Overview

This skill appears to be a coherent Word-document processing tool, with disclosed local file handling and setup steps but no artifact-backed evidence of hidden exfiltration, credential abuse, persistence, or destructive behavior.

This looks suitable for Word document workflows, but review the installer because it can add system packages, keep backups before edits or conversions, and avoid processing highly confidential documents unless you are comfortable with their contents being handled by the agent.

Findings (3)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Installing the skill may modify your machine by adding document-conversion software and Python packages.

Why it was flagged

The installer can add system packages and Python dependencies needed for document conversion. This is disclosed and aligned with the skill, but it changes the user's environment.

Skill content
sudo apt-get install -y libreoffice ... install_sys_dep "pandoc" ... "poppler-utils" ... "antiword"
Recommendation

Review the install script before running it, install in a controlled environment if possible, and only proceed if you are comfortable adding those dependencies.

What this means

The agent can make changes to Word documents when you ask it to edit or convert them.

Why it was flagged

The skill intentionally exposes local document editing and low-level XML packing/unpacking workflows. This is central to the purpose, but users should recognize it can alter files.

Skill content
编辑/修改  → edit_docx.py(JSON 规则)或 XML 编辑(office/unpack → 修改 → pack)
Recommendation

Keep backups of important documents and review generated or edited files before relying on them.

What this means

Sensitive text, tables, images, or metadata from documents you ask it to analyze may be processed by the agent.

Why it was flagged

The skill is designed to extract document contents and may use the agent/model to summarize or analyze them. This is expected for the feature set, but document contents may enter the agent context.

Skill content
读取 | 提取文本、表格、图片、元数据 ... 分析 | 文档摘要、关键词提取、字数统计 | word_count.py + AI
Recommendation

Use the skill only on documents you are comfortable sharing with the agent, and avoid unnecessary analysis of confidential files.