Extending Pulumi Neo: MCP Servers and Cloud CLIs

Infrastructure work lives between browser tabs: PagerDuty in one, Datadog in another, the AWS console in a third, your Pulumi state somewhere else entirely. Pulumi Neo already understands your infrastructure — your code, your stacks, your state. Its newest integrations extend that reach in two directions: into the third-party systems your team uses to plan and observe, and out to the cloud CLIs that actually drive your infrastructure.

In this live workshop, we’ll put the integrations to work on the problems they were built for: an alert that needs triage, a metric history that explains it, live cloud state that doesn’t match what the code says, and a fix that should ship as a reviewed pull request rather than a console click. You’ll see how Neo pulls context from the tools your team already uses — MCP integrations for Atlassian, Datadog, Honeycomb, Linear, PagerDuty, and Supabase — and reaches live cloud state through scoped CLI integrations for aws, gcloud, az, and kubectl, with each one configured, scoped, and controlled per task.

We’ll close with the newest additions to the integrations story: CLI integrations inside scheduled Neo Automations, so your recurring checks can reach live cloud state without a human in the loop, and Neo in your editor — Zed, JetBrains, VS Code, or Cursor — where it inherits the CLIs you’re already authenticated to.