The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes untrusted external content. * Ingestion points: Extracts text content from external URLs and uploaded documents in the Content Analysis Workflow (SKILL.md). * Boundary markers: None. The instructions lack delimiters or explicit directives to treat the crawled content strictly as data, increasing the risk that the model might follow instructions found within that content. * Capability inventory: Uses specialized vendor tools (analyze_google_search_results, analyze_related_search_queries, search_queries_vs_search_results, generate_seo_report) for analysis. No high-risk capabilities like shell access or arbitrary code execution were detected. * Sanitization: None. There is no evidence of filtering or sanitization of the external data before it is presented to the agent.

seo-analysis