SkaleAgents

The SkaleAgents homepage

SkaleAgents is a control plane for AI DevOps agents. It connects the AI you already use to specialist reviews of your application and your infrastructure, then tells you what to change. The findings come back in the same conversation, so your editor stays the starting point.

Getting started takes three steps. First, add SkaleAgents to your client: choose your client on the MCP page and add the hosted server URL. It works with Cursor, Claude Code, Claude Desktop, ChatGPT and Codex. Second, connect through Google OAuth and approve MCP access; credentials are handled automatically. Third, ask your assistant for an architecture review with the relevant source or infrastructure and read the findings right there.

Reviews cover security, DevSecOps, reliability, AI workloads, MLOps, scaling and cost, and performance. A useful review connects a finding to the code behind it. For example, bring a Terraform snippet or a Kubernetes manifest and ask for exposed services, broad permissions and the smallest useful fix, such as SSH open to 0.0.0.0/0 or a publicly accessible database.

You can explore without an account. Try the editable demo, which updates sample findings in your browser as you change deployment settings like replicas, readiness probes and memory limits. Browse the public bot directory to find a specialist for your next review, open the prompt library and copy a prompt into your assistant, read what each review check looks for, or use the free checklists in your own assistant.

Sign in with Google when you want to save your own work, publish specialist agents and connect your tools. Public bots, prompts and review checks are free. Starter, Pro and Max organization plans add organization audits and scheduled monitoring when a team needs a repeatable review process.