xcode-26
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documentation covers frameworks that ingest external data (LLM sessions and WebViews), which represents a potential attack surface for apps using these APIs.
- Ingestion points:
LanguageModelSessioninFoundationModels-Using-on-device-LLM-in-your-app.mdandWebViewinSwiftUI-WebKit-Integration.md. - Boundary markers: Suggested use of session instructions to steer AI behavior.
- Capability inventory: Support for tool calling and script execution.
- Sanitization: Documentation focuses on implementation rather than data validation.
- [EXTERNAL_DOWNLOADS]: The skill mentions the official
apple/app-store-server-library-swiftrepository on GitHub for transaction signing, which is a trusted developer resource. - [DYNAMIC_EXECUTION]: The documentation for
WebKitintegration includes standard usage examples forcallJavaScriptto interact with web content.
Audit Metadata