i4h-workflow-validate
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill clones the
i4h-workflowsrepository fromhttps://github.com/isaac-for-healthcare/i4h-workflows. This is a vendor-owned resource managed by the Isaac for Healthcare team at NVIDIA. - [REMOTE_CODE_EXECUTION]: The skill executes the
./run.shscript and other tools from the downloaded repository to perform simulation rollouts. The source repository URL can be overridden through theI4H_WORKFLOWS_REPO_URLenvironment variable. - [COMMAND_EXECUTION]: Shell commands are used for environment resolution, repository management, simulator lifecycle control, and inspection of rollout results.
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources including simulation logs, JSON manifests, and HDF5 recordings, which creates an attack surface for indirect prompt injection.
- Ingestion points: Shell output from
./run.sh list,run.jsonmetadata, andverify.hdf5simulation recordings. - Boundary markers: No specific delimiters or warnings for external data ingestion are defined in the instructions.
- Capability inventory: File system access (git clone, cd), shell execution (run.sh, stop.sh), and dynamic execution via
uv run. - Sanitization: No specific sanitization or validation of ingested simulation data is described beyond verifying the exit status and success flags.
Audit Metadata