Wouldn't that issues list be a little less intimidating if you had some help from Sentry? We're shipping our Sentry Agent for Linear, and you can use it to call Seer and get Root Cause Analysis and Solutions back. Check it out!
It turns out, Sentry does A LOT these days. Errors, Logs, Replays, Traces, and even Agent and MCP monitoring… but all this observability works better with a good foundation. We’re going to step back to 0, and show you how to build a “not bad” Sentry implementation from the ground up.
Retail stores have long relied on a secret weapon to measure and improve the shopping experience: the secret shopper. Posing as ordinary customers, they evaluate the customer experience, spotting friction points like hard-to-find items, gauging the quality of customer service, and testing how seamless the checkout process feels.
Whether you’re building agents in your applications, or standing up an MCP server because its the new cool thing, working with AI is just different. Trying to figure out why it does the weird things it does is hard.
See how you can started using MCP Server Monitoring against your MCP Servers REALLY fast using using Sentry. One line to wrap the code, and you're good to go. Check out this video where we set it up using Vercel's mcp-handler in Next.js.
This could’ve been prevented. This should have been prevented. This too. We all hate getting tagged in PRs. The time, the blame for when you inevitably miss something, and constant “I wouldn’t have written it that way” feeling is just hard to shake. LLMs promised this would get easier. Promised they would do it for us. But as we’ve seen, we’re not there yet. But this is what Sentry does for a living. We catch bugs… in prod.
Dan Mindru is a Frontend Developer and Designer who is also the co-host of the Morning Maker Show. Dan is currently developing a number of applications including PageUI, Clobbr, and CronTool. I find it remarkable that we’re getting so many AI startups every day. As software engineers, most of us like to know what our software is actually doing. We plan, review, and perform automatic tests to verify it’s working as expected. Then we do a round of manual testing for good measure. Not with AI.