email-marketing-bible-skill
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and audit user-provided campaign data and email copy, creating an inherent surface for indirect prompt injection.
- Ingestion points: Audit requests, campaign reviews, and deliverability diagnosis prompts in SKILL.md.
- Boundary markers: The skill does not suggest the use of specific delimiters (like XML tags or markdown blocks) to isolate untrusted user content from instructions.
- Capability inventory: Relies on standard agent analysis; no specialized tool invocations (like subprocess or file-write) are explicitly requested in the instructions.
- Sanitization: No mention of sanitizing or escaping user-provided HTML or text content.
- [EXTERNAL_DOWNLOADS]: The skill references external resources including the installation repository on GitHub and a supporting website (emailmarketingskill.com). These are documented as informational resources and are not fetched or executed automatically during skill operation.
Audit Metadata