Frequently Asked Questions

ops-agent is an AI-powered operations teammate that lives in your team channels. It diagnoses issues, runs health checks, and executes routine ops tasks through natural conversation — no commands to memorize.

Unlike general-purpose AI chatbots, ops-agent is purpose-built for operations. It understands your infrastructure context, can execute multi-step diagnostic workflows, and integrates directly with your existing DevOps tools and channels.

ops-agent connects to your infrastructure via PPIO sandbox — a secure, isolated execution environment. It supports standard Linux tooling, cloud CLIs (AWS, GCP, Azure), Kubernetes, Terraform, Ansible, and popular monitoring platforms.

Yes. Each user gets a private, isolated PPIO sandbox. Your credentials, server access, and conversation history are never shared with other users. Data is encrypted in transit and at rest. We do not use your data to train models.

Data is stored in US-East (Virginia) by default. EU region coming in Q4 2026.

Sign up for free (no credit card needed), connect a supported team channel, and start chatting with your agent. Average setup time: under 2 minutes.

Your agent pauses and notifies you. Free tier users wait until the next monthly reset. Pro/Team users can top up anytime — we never auto-charge you.

Minimal channel permissions: respond to messages, receive mentions or commands, and read the conversation context needed for diagnostics. We never request admin or workspace-management permissions unless a future integration explicitly requires it and you approve it.

Yes, anytime from Settings. Your credits remain active until the end of the billing period. No questions asked.

Yes! The Free tier includes 100 credits/month forever — enough for about 50–100 conversations. No credit card required.

Pro: 99.5% uptime SLA. Team: 99.9% uptime SLA with priority incident response. Free: best-effort.

ops-agent complements — not replaces — your existing tools. It's the conversational layer on top: instead of navigating 5 dashboards during an incident, you describe the problem and the agent pulls data from all of them.