reply-classification
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies and describes a workflow for processing untrusted inbound email replies, which serves as a significant attack surface for indirect prompt injection.
- Ingestion points: The skill analyzes raw email body and subject content received from external senders (as described in SKILL.md).
- Boundary markers: The documentation explicitly recommends the use of structured prompts to separate instructions from the untrusted email content.
- Capability inventory: There are no scripts, tools, or active capabilities defined within this skill; it is entirely instructional.
- Sanitization: The guide provides specific recommendations for sanitizing email content before passing it to language models to mitigate potential injection attempts.
Audit Metadata