Start system-problem investigations from a chat command
PostNL's AI team built a Slack command that has AWS DevOps Agent investigate system problems, so on-call engineers spend less time gathering evidence while keeping the decisions.
Done with AWS DevOps Agent
What they did
PostNL's AI team set up AWS DevOps Agent with limited access permissions. Engineers type a Slack command such as '/watchdog investigate the 429s we are seeing'. That kicks off an investigation of the system's logs, performance numbers and how its parts connect. The findings come back in the same Slack thread.
How it went
The author says the agent now does the first round of investigation. On-call engineers spend less time collecting evidence and still decide what it means and what to do. No figures are given.
Worth knowing
Much of the setup could be written as code, but the agent's skills still had to be set up by hand in Amazon's web console at the time.
Try it yourself with AWS DevOps Agent
Help me set up an incident-investigation workflow: when I post '/investigate [symptom]' in [Slack channel], gather evidence from [our logs and metrics source], recent deployments and service dependencies for [service], then post a hypothesis with supporting evidence to the thread. Use only read-only access scoped to [account or resources]. Show me the architecture and permissions plan first and wait for my approval before deploying anything.
Swap the highlighted [parts] for your own details.
Source: DEV Community · Matheus das Mercês · Oct 15, 2026
Written with AI assistance from the original post and checked against our editorial guidelines. Results are as reported by the source, not independently verified.
