med-international
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No executable code, scripts, or configuration files capable of running commands were found. The skill consists exclusively of Markdown documentation and reference materials.
- [SAFE]: External links point to reputable international bodies (e.g., UN, UNHCR, IMF, WHO), official government ministries (Taiwan MOFA, EU EEAS, French and German Foreign Ministries), and established journalism safety organizations (CPJ, Dart Center). No malicious redirects or typosquatted domains were detected.
- [SAFE]: No evidence of obfuscation or hidden instructions. Multilingual content (Chinese characters) and fictional language examples (using characters like 'չ') are used strictly for contextual training on translation sensitivity and geopolitical framing.
- [SAFE]: The skill does not request or perform any network operations, file system access, or credential harvesting. It focuses entirely on text analysis and generation logic.
- [SAFE]: Indirect prompt injection risk is negligible as the skill lacks executable capabilities or tool access. It incorporates rigorous source-chain auditing steps which serve as a safeguard against misinterpreting or blindly trusting input data.
Audit Metadata