seo-content-engine
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection due to its core function of processing external web content.
- Ingestion points: External data is ingested from the provided company-url and through competitor research mentioned in the audit steps.
- Boundary markers: There are no explicit instructions or delimiters used to separate external content from the agent's internal instructions or to warn the agent against executing embedded commands within fetched data.
- Capability inventory: The skill orchestrates multiple sub-skills, including content-asset-creator and seo-content-audit, which are capable of writing files and conducting automated web research.
- Sanitization: No validation or sanitization mechanisms are described for the data fetched from external sources before it is processed by the agent.
Audit Metadata