What's Changed
- .github/workflows: install the cli with the setup action by @lukasmalkmus in https://github.com/axiomhq/axiom-go/pull/460
- chore(deps): bump github.com/sirupsen/logrus from 1.10.0 to 1.10.1 by @dependabot[bot] in…
npx @buildinternet/releases get axiomMinimum Go version raised to 1.26, and integration tests no longer leak datasets.
LimitError can now be inspected as an HTTPError, simplifying error handling for users who previously had to type-assert or handle it separately.
An AI agent can now create a standalone Axiom organization, work in it for 24 hours, and hand it to a human via a claim link. Multi-query and dual Y-axis charts extend beyond metrics to structured, APL, and MPL queries, and legends gain aggregation columns and sorting.
Large dashboards can now be divided into collapsible sections that group related elements, and queries from collapsed sections stop running.
Gauge chart elements can now be added to dashboards and configured with data, including value fields, min/max ranges, thresholds, and custom styling.
Dataset schema locking makes a dataset's schema immutable, blocking new columns, column removal, renames, type changes, and other structural modifications at the API level.
The Axiom Grafana data source now runs both APL and MPL queries with a unified v2 query model, CodeMirror language support, and automatic migration of saved queries. APL gains four weighted aggregations, the MCP server adds monitor and notifier management, and metrics datasets get a prebuilt Kubernetes dashboard on arrival.
Hyper-cardinality, unified with logs and traces, and fully queryable by AI agents through MCP and a dedicated metrics skill.
Dashboards API gives you full programmatic control over dashboards.
Turn your AI coding agents into evaluation suite authors with the Write Evaluations skill.
Score your AI capability's outputs against live production traffic in real-time.