harness-setup

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: Uses the Bash tool to perform project initialization, directory creation, and file system management for environment setup.
  • [EXTERNAL_DOWNLOADS]: Installs the @openai/codex package globally via NPM, which is an expected dependency from a well-known vendor for this skill's functionality.
  • [COMMAND_EXECUTION]: Configures shell script hooks (pre-tool.sh, post-tool.sh) which are designed to integrate with the agent's execution lifecycle.
  • [COMMAND_EXECUTION]: Implements maintenance routines that use standard shell utilities to prune old log files and rotate agent trace data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 11:42 AM
Security Audit — agent-trust-hub — harness-setup