product-marketing-context
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface by processing external data from scraped web pages, reviews, and emails. It mitigates this risk with explicit instructions to the agent to treat such data as untrusted and to ignore any embedded commands. Evidence of ingestion is found in SKILL.md and references/skill-contract.md. The skill relies on descriptive logic rather than specific structural markers for boundaries. Capability inventory includes listed marketing tool integrations in references/connectors.md. Sanitization is handled through the instruction to disregard external instructions.
- [SAFE]: The skill follows established security principles, requiring explicit user authorization for external side effects and persistent writes. It also provides clear guidance to avoid hardcoding credentials in the skill or source control.
Audit Metadata