The Agent Skills Directory

[SAFE]: The skill implements secure document processing workflows. XML manipulation is primarily handled using the defusedxml library to protect against XML External Entity (XXE) and other XML-based attacks. Subprocess calls for document validation and diffing in ooxml/scripts/pack.py and ooxml/scripts/validation/redlining.py use argument lists, preventing shell command injection. The ingestion of untrusted document content (Category 8 surface) is a core part of the skill's purpose and is handled defensively. No evidence of prompt injection, data exfiltration, or obfuscation was found.

docx