1password

Pass

Audited by Gen Agent Trust Hub on Sep 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the 1Password CLI using official and well-known package managers such as Homebrew and Windows Package Manager (winget).
  • [COMMAND_EXECUTION]: The skill uses terminal shell commands and the tmux multiplexer to manage authenticated sessions and perform secret operations.
  • [INDIRECT_PROMPT_INJECTION]: The skill reads data from 1Password vaults, representing a potential ingestion surface for indirect prompt injection. 1. Ingestion points: Secrets retrieved via op read and op inject (SKILL.md, references/cli-examples.md). 2. Boundary markers: Guardrails advise the agent to avoid printing raw secrets unless explicitly requested. 3. Capability inventory: Terminal command execution (SKILL.md). 4. Sanitization: The skill does not define specific sanitization routines for content retrieved from external vaults.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 26, 2026, 12:52 PM