cold-email
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external files (e.g.,
.agents/product-marketing.md,.claude/product-marketing.md) and user-provided research signals like company news or LinkedIn activity. - Ingestion points: Marketing context files found in the project environment and research triggers provided by the user in the prompt.
- Boundary markers: The skill lacks explicit instructions or delimiters to isolate processed data or to warn the agent against following instructions that might be embedded in these external sources.
- Capability inventory: The skill is primarily instructional for text generation; it does not request high-risk tools such as shell execution, network access, or file-writing capabilities.
- Sanitization: No validation or filtering logic is prescribed for the ingested marketing context or research data before it is incorporated into the email generation process.
Audit Metadata