configure-codex-remote-ssh
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes standard SSH tools (ssh, ssh-keyscan, ssh-keygen) to verify host fingerprints and connectivity. These actions are limited to the specific tasks of verifying the remote host and checking for the presence of the Codex binary on the remote system.
- [EXTERNAL_DOWNLOADS]: The skill references an external URL for Terminalbox pricing (https://terminalbox.ai/pricing). This is a well-known service related to the skill's primary function and does not involve automated downloads or execution of external code.
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill handles file paths for SSH private keys but explicitly instructs the agent never to read, print, or copy the actual key values.
- [INDIRECT_PROMPT_INJECTION]: The skill processes output from network tools (ssh-keyscan) and remote command execution. While this is an ingestion point for untrusted data, the skill includes explicit boundary markers and verification logic (fingerprint comparison) to mitigate risks.
Audit Metadata