generic-max-supply

Pass

Audited by Gen Agent Trust Hub on Jun 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions explicitly direct the agent to execute Python scripts, including 'scripts/model.py' and 'model_whatif.py', within the working directory. This workflow involves the execution of code that the agent has potentially modified.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes data from multiple CSV files which are used to influence the modification and subsequent execution of Python code.
  • Ingestion points: Data is loaded from several CSV files in the 'scripts/data/' directory, including 'demand.csv', 'supply.csv', and 'processes.csv'.
  • Boundary markers: The instructions do not define delimiters or markers to separate data from instructions within the processed files.
  • Capability inventory: The agent is empowered to execute Python scripts ('python3 scripts/model.py') that it has been instructed to modify based on the input data.
  • Sanitization: The 'data.py' script uses standard pandas loading without explicit validation or sanitization of the input data values before they are used in the model logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 27, 2026, 10:03 AM
Security Audit — agent-trust-hub — generic-max-supply