vietnamese-business-comms

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The scripts/validate_copy.py script employs a plugin system to load business-specific rules. It uses the importlib.util module to dynamically find and execute Python files matching the rules_*.py pattern within its own directory. While this is an architectural choice for extensibility, it involves runtime code execution from computed paths.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection as it is designed to ingest and process user-provided Vietnamese copy. 1. Ingestion points: Files passed to scripts/validate_copy.py for validation. 2. Boundary markers: The skill uses comment-based markers like <!-- proof: ... --> for directive suppression. 3. Capability inventory: The skill can read/write files and execute local Python scripts. 4. Sanitization: The scripts use regex for pattern matching but do not perform explicit sanitization of the input text.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 07:25 AM
Security Audit — agent-trust-hub — vietnamese-business-comms