ce-test-xcode
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill introduces a potential indirect prompt injection surface because it processes untrusted application and simulator log outputs.
- Ingestion points: Simulator logs containing exceptions, crashes, and network request details are ingested during the execution phase in
references/test-and-report.md. - Boundary markers: There are no explicit boundaries or instructions telling the model to ignore natural language commands embedded within the logs.
- Capability inventory: The skill is capable of invoking external tools/skills (
ce-debug) and executing local simulator utilities (xcrun simctl openurl). - Sanitization: No content filtering or validation is performed on the log stream before it is processed by the agent.
Audit Metadata