The Agent Skills Directory

[SAFE]: The skill provides structured instructions for SEO analysis without introducing malicious code, persistence mechanisms, or unauthorized data access. It correctly identifies technical limitations of fetching tools and directs users to official resources from trusted organizations like Google.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to audit external websites using tools like web_fetch. This creates an attack surface for indirect prompt injection where a malicious website could contain instructions intended to influence the agent's behavior. However, the skill's primary focus is on objective technical analysis, and it highlights that the platform's fetching tools provide a layer of sanitization by stripping script tags.
Ingestion points: External website content retrieved via web_fetch or browser-based inspection.
Boundary markers: The skill does not explicitly define delimiters or markers to separate untrusted site content from system instructions.
Capability inventory: The skill utilizes data retrieval tools (web_fetch) and suggests using browser developer tools for DOM inspection.
Sanitization: The skill explicitly mentions that web_fetch strips <script> tags, which serves as a security control against direct script execution during the audit process.

seo-audit