The trace waterfall now includes time compression, pinned attributes, and clickable Web Vital pills to help you focus on the parts of a trace you want to investigate.

Resolve, reopen, or mark User Feedback as spam, download Session Replays as rrweb JSON, and list a project's DSNs from the Sentry CLI.

View or explain several issues at once, see event attachments, and split WebAssembly debug data from the Sentry CLI. Older traces and large project lists no longer fail.

Agents using Sentry MCP now search each dataset with a dedicated tool, can see which events Sentry dropped and why, search logs with regular expressions, and open dashboards from a URL.

You can now filter and sort conversations in Explore > Agents on totals like cost, token count, tool calls, and errors.

The React Native SDK now records the calls, duration, and errors of native modules, and adds the active call to crash reports.

The Android SDK can now instrument Nav3 back stacks to capture navigation transactions, breadcrumbs, screen names, and more.

The Rust SDK can now report native crashes, such as segfaults in unsafe code, to Sentry.

Sentry Logs search now supports regular expressions (regex), so that you can search log lines that match a pattern, not just by the exact text.

The JavaScript SDKs now trace more AI frameworks, providers, and classifiers across Node.js, Bun, Cloudflare, and Deno, with more consistent data in traces.

Generate custom dashboards with an LLM agent.

You can now add rich text markdown widgets to your custom dashboard.

Uptime monitors now support custom configuration, so you can define what a successful check actually looks like.

Metrics, currently in open beta, now support creating alerts.

We've overhauled how custom dashboards render chart legends.

We've added Tombstone support for enhanced Native crash reporting on Android

Send Metrics using our .NET and Unity SDKs

Send logs using the Kotlin Multiplatform SDK

Size Analysis is now generally available in Sentry. It helps you monitor and reduce your mobile app size so you can catch regressions before they reach end users.

We've renewed first-party support for data forwarding and migrated users onto organization-level configuration. You can now forward your errors across projects to Splunk, Segment and Amazon SQS with a single settings change.