gmail
Pass
Audited by Gen Agent Trust Hub on Apr 24, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits vulnerability to indirect prompt injection (Category 8) by ingesting untrusted data from external sources and providing access to sensitive tools. \n
- Ingestion points: The tools
read_email,list_emails,search_emails, andget_email_threadfetch content from the user's Gmail mailbox, which contains messages from external, untrusted senders (SKILL.md). \n - Boundary markers: There are no instructions or delimiters defined to separate untrusted email content from the agent's core instructions, increasing the risk of the agent obeying commands embedded in emails. \n
- Capability inventory: The skill provides powerful write and delete capabilities, including
send_email,delete_email, andbulk_delete_emails, which could be abused if an injection is successful (SKILL.md). \n - Sanitization: The documentation does not describe any sanitization or validation of the ingested email bodies or subjects to prevent instruction injection.
Audit Metadata