On demand
Cooking with Claude: Building an SRE Incident Response Agent
Join Gagan Bhat from Anthropic's Applied AI team for a live session on how to build an SRE incident response agent with Claude Managed Agents
In this session, you'll wire it together end-to-end, live in the terminal. We’ll show how a webhook kicks off the agent with a single API call, how a Skill teaches it your team's runbook conventions, and how custom tools let it open a PR and gate that action behind a human's final call - all with full observability in the Console.
About Cooking with Claude
Cooking with Claude is a series of live sessions built on Claude Cookbook, a free, open library of practical guides and runnable examples for building with Claude. Each one is a complete, working build written by Anthropic's Applied AI team (and community contributors) to show you exactly how something is done, end to end, in code you can run and adapt. In our live sessions, the authors walk you through a build and answer your questions in real time.
What you’ll learn
Upload a Skill and attach it to a Claude Managed Agent
Mix the built-in toolset with custom tools your own application handles
Start a session from a webhook payload
Gate a destructive action behind human approval
Read the full session trace in the Console