sg-home-chef-skill

Pass

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The evaluation and test scripts (evals/grade.py and tests/test_sg_home_chef.py) utilize subprocess.run to execute the skill's internal logic script (scripts/sg_home_chef.py). These executions are restricted to the local environment and are used for functional validation.
  • [SAFE]: The core script scripts/sg_home_chef.py is designed with a zero-dependency approach using only Python standard libraries. It does not perform network operations, exfiltrate data, or access sensitive system configuration files.
  • [SAFE]: The instruction set in SKILL.md provides a clear operating framework with specific constraints on pricing and vendor speech, reducing the risk of the agent providing unverified or misleading information.
  • [SAFE]: Input sanitization is implemented in the Python script to normalize user-provided dish names using regular expressions, which helps protect against malformed inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 30, 2026, 01:49 PM
Security Audit — agent-trust-hub — sg-home-chef-skill