aiui-debug
Pass
Audited by Gen Agent Trust Hub on Nov 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes logs and source code (Ingestion points: SKILL.md, references/device-and-logs.md). Boundary markers are established by instructions to separate facts from inferences. Capability inventory includes ADB commands for diagnostics. Sanitization is enforced via mandatory redaction of credentials and PII from all outputs.- [COMMAND_EXECUTION]: The skill uses standard Android Debug Bridge (ADB) commands for log capture and system state inspection. These are used for legitimate diagnostics and are accompanied by instructions to respect device authorization and avoid unnecessary modifications.
Audit Metadata