The Agent Skills Directory

Infrastructure Management via CLI: The skill utilizes the vercel CLI to perform administrative tasks such as adding WAF rules, blocking IP addresses, and configuring system bypasses. These operations are essential for the skill's purpose of platform-level security management.
Staged Change Management: The skill follows a safe-by-default pattern where most configuration changes (rules and IP blocks) are staged as drafts. It explicitly instructs the agent to review changes with the user and requires a manual vercel firewall publish command to take effect, ensuring human-in-the-loop oversight for production modifications.
Service-Level Access Controls: Several high-impact commands, such as Attack Mode and pausing system mitigations, are correctly identified as requiring interactive confirmation and are flagged as blocked for automated agents. This prevents the agent from making emergency-level changes without direct user intervention.
Project Analysis for Configuration: The skill suggests scanning the project repository for API endpoints to recommend rate-limiting rules. While this involves processing untrusted codebase data, the skill mitigates risk by recommending a 'log-only' mode first followed by human review of the generated rules.
Trusted Platform Integration: All external references, documentation links, and CLI tools are directed toward official Vercel resources, representing standard integration with the vendor's own infrastructure.

vercel-firewall