swiftui-uikit-interop
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides documentation links to the external domain
sosumi.ai(e.g.,https://sosumi.ai/documentation/swiftui/UIViewRepresentable). These function as supplementary reference links for platform APIs. - [DATA_EXFILTRATION]: Code recipes in
references/representable-recipes.mdinclude examples usingURLSession.shared.data(from: url)to fetch external PDF content. This is standard network functionality consistent with the skill's purpose as a developer resource. - [INDIRECT_PROMPT_INJECTION]: The skill documents methods for ingesting and rendering external data from URLs, which represents a potential attack surface for indirect prompt injection via malicious document content.
- Ingestion points:
references/representable-recipes.md(RemotePDFViewrecipe fetching PDF data from a remote URL). - Boundary markers: None; the documentation assumes the developer provides trusted URLs.
- Capability inventory: Includes network fetching via
URLSessionand rendering viaPDFViewinreferences/representable-recipes.md. - Sanitization: Standard error handling is implemented for the network request fetching the data.
Audit Metadata