AI Agents for Ops
Learn where AI coding agents fit into server operations, from routine maintenance to running an agent continuously on a VPS. You’ll be able to choose what to delegate, scope credentials, connect tools through MCP and set guardrails for production access.
The AI Ops Playbook: What to Hand Your Agent, What to Keep
Decide which server tasks to give Claude Code, where human approval belongs, and how to turn agent work into dependable automation on a single VPS.
All guides in this topic
Size a VPS for Always-On Agents: OpenClaw, Hermes or n8n
Choose a VPS for OpenClaw, Hermes or n8n. Size CPU, RAM and storage for API calls, browser jobs and workflows, with practical checks before going live.
Secrets and Scoped API Tokens for AI Agents
Keep production secrets out of agent sessions. Choose scoped API tokens, isolate credentials on Ubuntu 24.04 LTS, and verify access before allowing writes.
Claude Code on a VPS vs SSH From Your Laptop
Choose whether to run Claude Code on your laptop or an Ubuntu VPS, with practical advice on SSH access, persistent sessions, permissions and production risk.
Guardrails for AI Agents on Production Servers
Set practical guardrails for Claude Code on production servers using restricted Linux accounts, blocking hooks, scoped credentials, and reviewed deployments.
Why Your AI Agent Should Talk to Cloudflare, With Limits
Connect Claude Code to Cloudflare MCP with scoped permissions. Choose read access, control production changes, and verify access limits before relying on it.