image-seo
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts a website URL as an argument to perform SEO audits. This creates a surface where the agent could ingest malicious instructions embedded in the target website's HTML, metadata, or image alt text.
- Ingestion points: User-provided URL specified in the
argument-hintfield. - Capability inventory: The skill is designed to analyze image properties (alt text, file names, structured data) which are controllable by external website owners.
- Boundary markers: No explicit delimiters or warnings to ignore instructions within the audited content are defined in the entry point file.
- Sanitization: No specific sanitization or filtering of external content is mentioned in the skill instructions.
Audit Metadata