The Agent Skills Directory

[SAFE]: The skill instructions and associated reference files do not contain any malicious patterns, such as credential theft, privilege escalation, or persistent access mechanisms. The logic is focused on providing expert analysis of website structure and content.- [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection because it audits data fetched from external websites using tools like web_fetch or curl. This is a necessary function for an SEO tool and does not indicate malicious intent from the skill author.
Ingestion points: Data from external URLs and local context files like .agents/product-marketing-context.md.
Boundary markers: Absent; the instructions do not explicitly provide delimiters or warnings to ignore directives found within audited web content.
Capability inventory: The skill is restricted to analysis and report generation; it lacks high-risk capabilities such as arbitrary shell execution, file system modification, or network exfiltration.
Sanitization: No explicit sanitization or filtering of external content is defined.

seo-audit