actionmailer-baseline

Pass

Audited by Gen Agent Trust Hub on Sep 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and suggests the installation of several well-known Ruby gems, including letter_opener, letter_opener_web, aws-sdk-rails, and mailgun-ruby. These are standard tools for local development testing and integration with reputable cloud email providers.
  • [INDIRECT_PROMPT_INJECTION]: The skill documents a surface for ingesting external data through provider webhooks (e.g., for Postmark) to handle email bounces and complaints. While this introduces an external data ingestion point, the skill follows best practices to mitigate risk. Ingestion points: The WebhooksController in SKILL.md processes external POST requests. Boundary markers: The instructions explicitly direct the developer to verify webhook signatures before processing data. Capability inventory: The skill uses ingested data to update user status in the database and manage global email suppression filters. Sanitization: Data is processed using standard ActiveRecord query methods which provide built-in protection against SQL injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 29, 2026, 02:26 AM