redline-docx-pl

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses uvx adeu to run the adeu engine and python for local helper scripts. These operations are restricted to the local environment and are intended for document processing.
  • [EXTERNAL_DOWNLOADS]: The skill downloads the adeu package using uvx from a standard registry (PyPI). This is a well-known service for package distribution and the tool's core functionality.
  • [PROMPT_INJECTION]: The skill includes instructions to strictly follow safety tiers (Read-only, Mutating, Destructive) and requires explicit human confirmation for destructive actions, which serves as a safeguard against accidental or malicious misuse.
  • [DATA_EXFILTRATION]: No network operations or data exfiltration patterns were detected. All document processing happens locally, and the sanitize command explicitly removes metadata to prevent accidental data leaks (RODO/GDPR compliance).
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 08:27 AM
Security Audit — agent-trust-hub — redline-docx-pl