product-marketing
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted data from repository files during the 'Auto-draft' workflow, creating an indirect prompt injection surface.
- Ingestion points: Repository files including README, landing pages, marketing copy, and package.json are analyzed to generate product context.
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing external file content.
- Capability inventory: The skill utilizes file system read access for repo analysis and write access restricted to the '.agents/product-marketing.md' document.
- Sanitization: No evidence of sanitization, escaping, or validation of the content read from the codebase before it is processed for document generation.
Audit Metadata