Agent Tracing for JavaScript now auto-instruments Anthropic, OpenAI, Vercel AI SDK, LangChain, LangGraph, and Mastra — including Jev classifier evaluations and MCP server monitoring — across Node.js, Bun, Cloudflare, and Deno.
See the full list 👇
https://t.co/eGFTjhfF0Y
@Siddharthxdev@sentry We have a dedicated dashboard for tools where you can also see per tool error rates. Go to Dashboards > Sentry Built > AI Agents Tool Details.
For per-tool redaction you can use before_send in order to target the tool and disable sending inputs/outputs: https://t.co/mmd3xA1fXu
Production traces contain rich feedback that rarely makes it back into training pipelines. Self-distillation methods allow us to close the continual learning loop and learn from every interaction.
We’ve added native support for OPSD and RMSD to our platform, AC2. Here’s a preview:
Another new Insight 😮 Caches and Queues – so you can finally figure out if your app is using caching correctly, why those cache misses are killing your latency, and see how queues are impacting your app's performance (while ensuring your background jobs aren’t being sent to /dev/null).
As usual, here’s the post with everything you need to get started: https://t.co/IHDSiHUwmL
@mvaled@getsentry Hi, @mvaled! We are planning to add metrics like the ones you mentioned in the future. We're currently focused on refining the product for performance metrics we already offer, as well as custom metrics.
@getsentry’s new metrics feature makes it easy to measure all sorts of things, like your function execution time in Python.
What’s more, we correlate your custom metrics with traces and other signals to help you identify the root cause faster!
Join here:
https://t.co/X1KxhAdBnl
You heard about it at Launch Week, and now it’s here. Metrics is available in alpha.
Track the data points you care about over time to pinpoint and solve issues with correlated traces.
🧵 👇
@moonrailgun@getsentry Oh, in that case you can find the list of feature flags to enable here: https://t.co/WPxKAPfPIo
Let us know what you think or if you run into any issues.