skills/nanocoai/nanoclaw/add-codex/Gen Agent Trust Hub

add-codex

Pass

Audited by Gen Agent Trust Hub on Oct 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPERSISTENCEDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses various commands to build, test, and manage the installation, including pnpm for task execution and launchctl or systemctl for managing the background agent service.
  • [EXTERNAL_DOWNLOADS]: It fetches content from a Git branch and installs the @openai/codex package from the official NPM registry, which are trusted sources for this provider's functionality.
  • [PERSISTENCE]: The installation adds the provider to the agent's manifest and configures it to run as a persistent service, ensuring it remains available after system restarts or container builds.
  • [DYNAMIC_EXECUTION]: The setup scripts programmatically spawn the Codex CLI to perform authentication tasks. This process includes security controls such as session isolation in temporary directories and the use of a credential vault.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 17, 2026, 08:21 PM