product-marketing
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection via its auto-drafting workflow.
- Ingestion points: As defined in references/guidelines.md, the skill scans several local file types including README, landing pages, marketing copy, and package.json to extract context.
- Boundary markers: The instructions lack explicit delimiters or markers to distinguish between the skill's system instructions and the untrusted data being read from the files.
- Capability inventory: The skill is empowered to perform local file reads across the codebase and write the resulting marketing document to .agents/product-marketing.md.
- Sanitization: No sanitization, validation, or filtering of the content read from the codebase is specified, allowing potential embedded instructions to influence the agent's behavior.
Audit Metadata