i4h-workflow-e2e
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill clones the primary workflow logic from a GitHub repository belonging to the Isaac for Healthcare organization (
https://github.com/isaac-for-healthcare/i4h-workflows). This is an expected operation for installing the necessary tools and scripts to run the pipeline. - [COMMAND_EXECUTION]: The skill executes shell scripts located within the cloned repository (e.g.,
run.sh,stop.sh). These commands are essential for driving the robot learning pipeline stages such as recording, replaying, and training. - [SAFE]: Analysis of all threat categories, including prompt injection and data exfiltration, confirms the skill is focused solely on its documented purpose of managing agentic workflows for robotics. It uses standard environment variable resolution and local file logging without suspicious network activity or credential access.
Audit Metadata