op-cli
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Data Exposure] (SAFE): The skill is specifically designed to prevent data exposure. It includes explicit warnings against printing secrets to stdout and provides secure alternatives such as piping directly to destination tools (e.g., wrangler, kubectl) or using subshells.
- [Command Execution] (SAFE): Uses standard CLI tools (op, kubectl, wrangler) and a built-in Python one-liner for JSON parsing. These are used as intended for secret management and deployment tasks.
- [Privilege Escalation] (SAFE): Includes the standard
eval "$(op signin)"pattern for 1Password session management. This is the documented method for authenticating the CLI and does not involve OS-level privilege escalation. - [Indirect Prompt Injection] (LOW): While the skill processes output from the 1Password CLI (item lists and labels), the risk is low as it uses structured JSON parsing and is used within a trusted workflow.
Audit Metadata