designing-email-templates
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates a read-modify-write workflow for email templates, creating a surface for indirect prompt injection where malicious instructions inside a template could influence the agent.
- Ingestion points: Fetches existing template design JSON via
workflows-get-email-templateas described in SKILL.md. - Boundary markers: Instructions lack explicit guidance to treat the fetched content as untrusted data or to use delimiters to prevent the agent from following instructions embedded within the template.
- Capability inventory: The skill possesses write capabilities through
workflows-update-email-templateandworkflows-patch-email-template, allowing it to persist changes influenced by potentially malicious template content. - Sanitization: No sanitization or validation logic is specified for the agent to perform on the retrieved JSON before incorporating it into its decision-making process.
Audit Metadata