dxe
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run standard development commands such as type-checking, automated tests, and production builds within the repository environment to verify interface quality. It also directs the collection of runtime diagnostics via platform-specific tools like Xcode, Gradle, and Logcat.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the target repository, including source code, UI components, and system logs. This creates a surface where malicious instructions embedded in the codebase or logs could attempt to influence the agent's findings or fix recommendations. However, the skill provides a structured rubric and explicit evidence protocols to mitigate the risk of acting on unverified data.
Audit Metadata