What's new this week

This section summarizes what was released this week.

Fast lane GA features

These features are generally available for all customers. Slow lane customers and tenants will have these features automatically enabled on August 26, 2026.

datasets: Dataset headers and the REST API now show real measured (p90) freshness instead of the configured goal interval, and the staleness reason reflects whether data is actually behind its goal (not just the threshold approximation). Eliminates confusion between "configured goal" and "actual data freshness" so engineers can trust freshness signals when diagnosing pipeline lag.
monitors: Alert context cards now show correlation tag values as clickable pivots — click any tag to jump to correlated logs, metrics, traces, or dashboards in the full alert scope. Turns static context metadata into direct navigation so engineers don't have to copy-paste tag values into separate searches.

Slow lane GA features

These features are enabled for all slow lane customers and tenants today.

ai-sre: Skill-based chart visualization generation for query cards.
monitors: The /v1/monitor-mutes REST API for bulk and programmatic mute management is now available to all customers.

The following features are also enabled for all slow lane customers and tenants today. These features were announced last week, but due to an issue in the slow-lane rollout automation, they were not actually enabled:

api: New REST API for Datasets. (This item was originally missing from and retroactively added to the June 24 changelog.)
monitors: Consecutive cron-scheduled alarms for the same condition stitch into a single ongoing alarm.
connections: CDC change tracking for source datastreams.

Public preview and private preview features

features are available by invitation only. features are available by inquiry.

ai-sre: AI-assisted monitor creation — the AI-SRE agent can now create anomaly monitors, read and edit notification action content, and prompt for changes when a monitor suggestion is declined. Available for AI-SRE customers only (gated by the AI-SRE master switch).

Bug fixes

The following issues are fixed in this week's release:

datasets: Fixed the label showing "Tags" instead of "Object tags" in the dataset header. (OB-60453)
datasets: Fixed a race condition that triggered dataset search before feature flags finished loading. (OB-63059)
monitor-v2: Fixed formula-expression monitors rejecting wildcard service bindings.
monitor-v2: Fixed timezone not being propagated to the evaluated data view. (OB-62826)
monitor-v2: Fixed the noDataRules field being omitted from the GET /v1/monitors/{id} response. (OB-61582)
monitor-v2: Fixed anomaly group filter errors when the schema is not yet loaded.
service-explorer: Fixed persisted group-bys being applied against stale schema after navigating back. (OB-61883)
dashboard: Fixed the "Set as default" checkbox being enabled for unsaved dashboards.
service-inspector: Fixed chart bounce on resize when snap-resizing the inspector panel.
apm: Fixed APM query endpoints returning 500 on client cancellation and query timeout; now returns 400 and 504 respectively.
nav-rail: Fixed navigation rail items flickering closed when moving the cursor between adjacent items.

REST API changes

The following changes are reflected in the Observe REST API as of this release.

New

ObjectTags (GET /v1/object-tags/search): Search object tag keys used across dashboards, datasets, worksheets, and other objects, with an optional case-insensitive substring filter. Results are sorted by object count descending, then alphabetically.
ObjectTags (GET /v1/object-tags/search/{key}): Search unique values for a specific object tag key, with an optional case-insensitive substring filter.

Terraform provider

The following updates are available in the Observe Terraform provider.

v0.14.61 (2026-07-27)

New features

resource_monitor_v2: Anomaly monitor configuration now uses typed algorithm fields, replacing the previous untyped schema.
All resources: entity_tags renamed to object_tags to match the updated API and UI terminology.

Coming soon

The following GA features are already available for fast lane customers, and will be enabled for slow lane customers and tenants next week.

apm: Global service dependency map accepts environment and namespace filters without selecting a focal service.
opal: ai_complete() accepts optional model_parameters, response_format, and show_details arguments.