natural-language
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill documentation and provided Swift code snippets accurately describe the use of Apple's NaturalLanguage and Translation frameworks for on-device linguistic analysis and text translation. Implementation patterns follow established development practices for these system APIs.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a processing surface for untrusted text data via
NLTokenizer,NLTagger, andTranslationSession. While it lacks explicit input sanitization or boundary markers in the provided code snippets, it correctly defines functional boundaries and delegates sensitive operations like OCR and generative AI to sibling skills, maintaining a standard security posture for text analysis tools. - [EXTERNAL_DOWNLOADS]: The skill includes reference links to documentation on
sosumi.ai. These are static informational links and do not involve remote code execution or automated software installation.
Audit Metadata