email-drafter
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides structured guidance for drafting emails. It does not contain any executable code, obfuscation, or dangerous system commands.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data (user-provided email threads and context) which could contain hidden instructions (Indirect Prompt Injection).
- Ingestion points: Untrusted data enters the agent context through pasted email threads and user-supplied facts as described in the 'What to ask for' section of SKILL.md.
- Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions embedded within the processed email threads.
- Capability inventory: The skill has the capability to write output to external systems, such as saving drafts to a notes app via a connector or creating email drafts in an email tool as specified in the 'Saving the draft' section of SKILL.md.
- Sanitization: There are no instructions for sanitizing or escaping the content of the email threads before they are processed or saved.
Audit Metadata