product-marketing
Pass
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via its automated drafting feature.\n
- Ingestion points: The agent is instructed to scan and read various files from the codebase (README, landing pages, marketing copy, package.json, and other docs) to generate marketing context.\n
- Boundary markers: The instructions lack explicit directives for the agent to treat file contents as untrusted data or to ignore embedded instructions during the auto-drafting process.\n
- Capability inventory: The agent can read arbitrary files within the repository and write to the
.agents/product-marketing.mdfile. The resulting document is explicitly designed to be used as a context source for other downstream marketing skills (multi-step chain risk).\n - Sanitization: There is no evidence of content sanitization or validation before the agent processes the information gathered from the codebase.
Audit Metadata