soia-env-codex-install

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/inspect_installation.py uses subprocess.run to verify the existence, version, and login status of the Codex CLI. These commands are diagnostic in nature and limited to verifying the tool's installation environment.
  • [EXTERNAL_DOWNLOADS]: The skill references official installation channels, including the OpenAI-owned chatgpt.com domain for shell scripts and the @openai/codex package from the npm registry. These are verified sources for the software being managed.
  • [REMOTE_CODE_EXECUTION]: The skill provides instructions for the agent to install the CLI using a remote shell script. This execution is performed from a trusted vendor's official URL and is necessary for the skill's primary function.
  • [DATA_EXPOSURE]: Security measures are in place to ensure that authentication tokens and API keys are not processed, stored, or logged by the skill, instead delegating authorization to the official browser-based flow.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 08:47 AM
Security Audit — agent-trust-hub — soia-env-codex-install