Getting logs into your LLM

/March 18, 2026

Field NotePremium3:55

Getting logs into your LLM

Make debugging product stuff easy by pulling the logs. Learn how the new Sentry CLI helps you read remote logs and issues locally for faster troubleshooting.

I've found the new Sentry CLI incredibly useful. I even aliased it under the old Sentry CLI name because the old one isn't good, but the new one is great. I keep my Sentry org and project in my .env.local file, and my authentication is at the user level. This way, machines running commands in the repo don't have to figure out the org and project.

I can use Sentry log list to see the logs, which is helpful for debugging production issues without giving access to the server. Sentry now has structured logs, allowing me to drain logs to Sentry alongside issues. This means everything flows from the production server into Sentry.

I can also use Sentry issues to view current issues. Sentry is hooked up to Slack, so I receive Slack messages for new errors. I can then use the CLI to pull the latest error and start diagnosing it. While some might prefer automated fixes, I still want to be the human in the loop.

The important thing is to have a way for my local machine to read remote logs and issues. This closes the loop between the production environment and my local machine where I do the work.