recipe-forward-labeled-emails

Pass

Audited by Gen Agent Trust Hub on Mar 20, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it retrieves untrusted content from email bodies and uses it to perform further actions without sanitization.
  • Ingestion points: Email message content is retrieved in Step 2 of SKILL.md using the gws gmail users messages get command.
  • Boundary markers: No delimiters or instructions to ignore embedded commands are present; the message body is directly interpolated into a new email template.
  • Capability inventory: The skill utilizes gws gmail +send in Step 3, providing a mechanism for data transmission.
  • Sanitization: There is no evidence of content escaping, validation, or filtering before the data is processed.
  • [DATA_EXFILTRATION]: The skill is designed to move sensitive email content from the user's inbox to an external email address. While this aligns with the stated purpose of a forwarding skill, it constitutes a movement of potentially private data to a fixed external destination.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 20, 2026, 04:57 PM