pva-build-test
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on shell commands for the build process and interaction with remote hardware targets. It uses
sshandscpto deploy and execute tests on Tegra boards. The instructions include parameter validation to mitigate risks associated with dynamic command assembly. - [PRIVILEGE_ESCALATION]: The skill describes the use of
sudofor administrative hardware configuration tasks on the target device. It explicitly recommends least-privilege configurations and warns against using unrestricted passwordless sudo, adhering to security best practices for embedded development. - [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific data and external reports, creating a surface for indirect prompt injection. To manage this risk, it utilizes strictly defined JSON schemas and templates for data handoff and validation, reducing the likelihood of the agent following instructions embedded in untrusted data.
Audit Metadata