skills/vm0-ai/vm0-skills/gmail/Gen Agent Trust Hub

gmail

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes shell commands (curl) and text processing utilities (printf, base64, tr, jq) to perform operations. These commands interact with Google's official Gmail API endpoints at gmail.googleapis.com to read, write, and delete user data.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes untrusted data from external email messages.
  • Ingestion points: Email content, subjects, and metadata are fetched from the Gmail API (documented in SKILL.md).
  • Boundary markers: Absent; there are no instructions provided to the agent to treat fetched email content as untrusted data or to use delimiters.
  • Capability inventory: The skill possesses significant capabilities, including reading emails, sending new messages, deleting content, and modifying account settings (filters, vacation replies).
  • Sanitization: No sanitization or filtering of the fetched data is performed before it is presented to the agent context.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 04:43 PM