amc-run-sample-calibration

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The Python script scripts/run_sample_calibration.py uses subprocess.check_call to initialize a Python virtual environment and os.execv to re-execute the script using the new environment's interpreter. This is a standard 'self-healing' pattern to ensure the requests library is available for API interactions without requiring global system changes.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface as it processes external files (JSON, PNG, ZIP) and reads microservice logs. However, the skill provides specific instructions to the agent to avoid fabricating results and uses these inputs solely for calibration logic within the defined project workspace.
  • [EXTERNAL_DOWNLOADS]: The skill references the NVIDIA ngc registry for downloading additional warehouse datasets. These are verified vendor resources intended for use with the AutoMagicCalib service.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 08:38 AM
Security Audit — agent-trust-hub — amc-run-sample-calibration