omh-physical-device-readiness
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is primarily designed as a safety-gating workflow for physical hardware operations. It explicitly prohibits claiming success for physical actions (like heating or moving a robot) without verified telemetry or operator approval.
- [COMMAND_EXECUTION]: The skill references the use of the
omhCLI tool (omh runtime record) to track workflow status. This command is part of the vendor's internal management system and does not interact with sensitive files or external networks. - [INDIRECT_PROMPT_INJECTION]: The skill defines ingestion points for external data such as user requests and device context in
SKILL.md. It mitigates potential injection risks by requiring human-in-the-loop approval for all physical actions and maintaining a strict "prepared-vs-observed" boundary in its reporting.
Audit Metadata