email-writing
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process user-provided data to draft emails and save them to the local file system, creating an attack surface for indirect prompt injection.
- Ingestion points: User input or context data used to fill templates in
SKILL.mdandreferences/email-templates.md(e.g., recipient names, email body, and follow-up notes). - Boundary markers: Absent. The instructions do not specify delimiters for user data or include directives for the agent to ignore instructions embedded within the processed content.
- Capability inventory: File system write access, as the skill provides a specific "Email Draft File Format" in
SKILL.mdand directs the agent to save drafts to files. - Sanitization: Absent. There is no guidance for validating or escaping content before it is written to the draft files, which could lead to the persistence of malicious payloads or control sequences.
Audit Metadata