mailchimp-automation
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a legitimate tool for Mailchimp automation using the Rube/Composio framework. It correctly identifies high-risk actions (sending campaigns) and instructs the agent to seek explicit user confirmation and perform testing before execution.
- [SAFE]: The inclusion of a Python code snippet for MD5 hashing is a standard requirement for the Mailchimp API (which uses hashed emails as identifiers) and does not involve any suspicious logic or external dependencies.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of untrusted data from external sources, creating a potential surface for indirect prompt injection.
- Ingestion points: The skill retrieves audience data, subscriber profiles, and campaign HTML content from the Mailchimp API (SKILL.md).
- Boundary markers: There are no instructions defining the use of delimiters or specific boundary markers to wrap data retrieved from tools before presenting it to the model.
- Capability inventory: The skill provides write access and communication capabilities, including tools to add/update subscribers and send email campaigns (SKILL.md).
- Sanitization: The instructions do not describe any sanitization or validation steps for content fetched from the remote API.
Audit Metadata