Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates processing of untrusted data from incoming emails, which could be used for indirect prompt injection attacks.\n
- Ingestion points: Fields such as
subject,text, andhtmlfrom theEmailobject inSKILL.mdare passed directly to the handler function.\n - Boundary markers: No delimiters or protective instructions are used to separate user data from agent instructions.\n
- Capability inventory: The skill has the capability to send emails using
std/emailand log output to the console.\n - Sanitization: The provided code does not show any sanitization or filtering of email content before processing.\n- [EXTERNAL_DOWNLOADS]: The skill fetches its core functionality from the vendor's module repository.\n
- Evidence:
import { email } from "https://esm.town/v/std/email";inSKILL.mdimports code from the vendor-associated domain.
Audit Metadata