The Agent Skills Directory

[SAFE]: The skill implements standard SEO diagnostic procedures. All tool interactions are limited to reading and analyzing publicly available website metadata and performance metrics. No unauthorized data access or persistence mechanisms were detected.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources, such as robots.txt files and HTML headers, which creates a potential surface for indirect prompt injection from malicious websites. 1. Ingestion points: tools like analyze_page, analyze_robots_txt, and analyze_sitemap used in AUDIT_WORKFLOW.md. 2. Boundary markers: No explicit delimiters or instructions to ignore embedded content are defined in the skill text. 3. Capability inventory: Diagnostic analysis only; no high-risk capabilities such as file-writing, code execution, or network exfiltration of sensitive files are utilized in the audit workflow. 4. Sanitization: No explicit sanitization of external content is defined. The overall risk is low due to the lack of dangerous downstream capabilities.

seo-technical-audit