harness-setup

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses Bash to automate the creation of project directories, configuration files, and symlinks required for the Harness v3 environment.
  • [COMMAND_EXECUTION]: Maintenance functions utilize shell utilities like 'find' and 'tail' to manage local logs and state files within the project's .claude directory.
  • [EXTERNAL_DOWNLOADS]: Includes instructions for installing the '@openai/codex' package from a well-known service and provides guidelines for the standard installation of Claude plugins to extend agent capabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 11:38 AM