geo-content-optimizer
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of existing product descriptions, marketplace listings, and buyer questions to perform content rewrites. This creates a potential surface where adversarial instructions hidden in the source content could attempt to influence the agent's behavior.
- Ingestion points: The skill collects "existing content, platform, verified product facts and policies" from external sources (SKILL.md).
- Boundary markers: The workflow requires recording an "evidence boundary" and identifying vague or unsupported facts before rewriting (SKILL.md).
- Capability inventory: The skill performs content transformation, structures data, and generates change logs.
- Sanitization: No explicit sanitization or escaping of the input content is described, though the "Domain rules" mandate factual accuracy and adherence to evidence.
- [EXTERNAL_DOWNLOADS]: The skill provides installation instructions using the
npx skillscommand to fetch theecommerce-seo-geo-skillspackage from the author's repository.
Audit Metadata