emails
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's design involves processing external context and user-supplied data to generate email sequences, which creates a surface for indirect prompt injection.
- Ingestion points: The agent is instructed to read from project-specific context files such as
.agents/product-marketing.mdor.claude/product-marketing.md(SKILL.md). It also processes user-supplied triggers, audience contexts, and conversion goals. - Boundary markers: The skill does not implement specific delimiters or 'ignore embedded instructions' warnings when interpolating external context into its output generation process.
- Capability inventory: The skill focuses on text generation. Analysis of the scripts and instructions reveals no use of tools for file system writes, arbitrary shell execution, or unauthorized network operations.
- Sanitization: There are no documented procedures for sanitizing or validating the contents of context files before they are processed by the agent.
Audit Metadata