docx-tracked-changes
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were identified. The skill is entirely descriptive and focuses on guiding the AI agent's text formatting during document redlining.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted document content provided by users. While this is an injection surface, the risk is negligible as the skill lacks any capabilities, tools, or shell access that could be exploited. 1. Ingestion points: User-provided document text or files (SKILL.md). 2. Boundary markers: None explicitly defined to isolate document content. 3. Capability inventory: The skill does not use any tools, commands, or network operations. 4. Sanitization: No sanitization of external text is specified.
Audit Metadata