The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The scripts/fetch_page.py script allows the agent to fetch content from web URLs. It implements Server-Side Request Forgery (SSRF) protection by validating that hostnames resolve to global IP addresses and blocking access to private, loopback, and reserved network ranges.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from web scrapes and third-party exports. To mitigate prompt injection risks, both SKILL.md and references/skill-contract.md include explicit instructions to treat such data as evidence only and to disregard any instructions embedded within processed content.
[COMMAND_EXECUTION]: The skill contains multiple utility scripts (scripts/seo_health_score.py, scripts/schema_generator.py, etc.) for data processing and report generation. These scripts use standard Python libraries for JSON handling and string manipulation without involving unsafe shell execution or dynamic evaluation of untrusted input.
[REMOTE_CODE_EXECUTION]: The scripts/generate_pdf_report.py script includes an optional dependency on the reportlab library. The script provides clear instructions for the user to install this standard package if PDF generation functionality is required.

seo-dataforseo