The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted HTML data from external websites, which creates a potential surface for indirect prompt injection. Ingestion points: Server-rendered HTML text retrieved via the web_fetch tool. Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands. Capability inventory: Actions are restricted to data observation and audit reporting. Sanitization: No content sanitization or validation methods are described for the fetched data.
[SAFE]: The skill instructions adhere to standard SEO practices and exclusively link to verified documentation from trusted technology providers. No evidence of malicious code, data exfiltration, or unauthorized system modifications was found.

seo-audit