The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill performs SEO audits by ingesting and analyzing external web content, which serves as a potential vector for indirect prompt injection. Ingestion points: The skill is designed to analyze live web pages, response headers, robots.txt files, and XML sitemaps (SKILL.md). Boundary markers: The instructions do not specify the use of delimiters or provide warnings to ensure the agent ignores instructions that may be embedded within the content it processes. Capability inventory: The agent utilizes the lighthouse_audit tool and general browsing capabilities to inspect external domains. Sanitization: There is no mention of sanitization or validation of the extracted web data before it is processed for audit conclusions.

seo