mailchimp-automation
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external Mailchimp accounts and possesses capabilities to send emails and modify subscriber lists, creating a surface for indirect prompt injection.
- Ingestion points: External data enters the agent context through tools like
MAILCHIMP_GET_LISTS_INFO,MAILCHIMP_LIST_MEMBERS_INFO,MAILCHIMP_SEARCH_MEMBERS, andMAILCHIMP_GET_CAMPAIGN_REPORT(documented inSKILL.md). - Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions that might be embedded within subscriber names, campaign titles, or email content.
- Capability inventory: The skill includes powerful write capabilities, including
MAILCHIMP_SEND_CAMPAIGN,MAILCHIMP_ADD_OR_UPDATE_LIST_MEMBER, andMAILCHIMP_SET_CAMPAIGN_CONTENT(documented inSKILL.md). - Sanitization: There is no evidence of sanitization, filtering, or validation of the external content before the agent processes it or uses it to generate new campaign content.
Audit Metadata