1password
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the
1password-cliusing Homebrew (brew), which is a well-known and trusted package management service.\n- [COMMAND_EXECUTION]: The skill executesopandtmuxcommands to manage authentication and secret retrieval. It usestmuxsessions to handle TTY requirements and capture output, which is a functional requirement for the interactive 1Password CLI.\n- [PROMPT_INJECTION]: Indirect Prompt Injection Surface:\n - Ingestion points: Data enters the agent context through
op read,op run, andop inject(File:references/cli-examples.md).\n - Boundary markers: The skill does not define specific delimiters for vault content or instructions to ignore embedded commands.\n
- Capability inventory: The skill can execute shell commands via
tmuxand manipulate environment variables usingop run.\n - Sanitization: There is no explicit validation or sanitization of the content retrieved from the 1Password vault before it is processed by the agent or shells.
Audit Metadata