model-usage

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/model_usage.py invokes the codexbar CLI tool using subprocess.check_output to fetch cost data. Arguments are passed as a list and the provider name is validated against a whitelist of expected values, which prevents shell injection.
  • [EXTERNAL_DOWNLOADS]: The skill metadata in SKILL.md includes instructions to install the codexbar CLI from a third-party Homebrew repository. This is a legitimate dependency required for the skill's functionality.
  • [DATA_EXFILTRATION]: The skill is documented to read local usage logs from configuration directories such as ~/.codex/ and ~/.config/claude/. This data access is necessary for the skill's purpose, and no patterns indicating the external transmission of this data were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 08:32 AM
Security Audit — agent-trust-hub — model-usage