translationese-cleaner-vi
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for text processing and linguistic analysis. It contains comprehensive instructions for identifying and correcting 'translationese' in Vietnamese text, with specific guidelines for preserving specialized terminology and legal structures. No malicious code, unauthorized network access, or data exfiltration patterns were identified.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input (Vietnamese text requiring cleanup). While it lacks explicit boundary markers for this input, the skill's capabilities are restricted to text manipulation without access to dangerous tools like shell execution, network operations, or file system writes.
- Ingestion points: User-provided text for cleanup (identified in SKILL.md).
- Boundary markers: None identified in the provided instructions.
- Capability inventory: No subprocess calls, network operations, or file-writing capabilities detected.
- Sanitization: None identified; however, the agent's logic is focused on linguistic transformation rather than instruction execution.
Audit Metadata