Core
More resilient long-running operations
0.6.10
Improves seed import status and Agent recovery during transient failures.
Fixed
nexia-seed:statusnow reports a stale heartbeat as a warning while an import remains running instead of returning a failed command status. Operators should confirm worker logs before retrying a long-running fixture.- Agent dashboards now keep their last successful data visible during a transient service failure, avoid duplicate session tabs after reload, and stop terminal transcript recovery after bounded retries.
- Agent input waits now keep showing elapsed time, and quantitative report-table values use locale-aware digit grouping.
- Document details now use the revision media returned by the API, preventing a blank screen after save. Saved edits appear immediately while document and revision history data refresh.
Upgrade required
- No SDK or App package release is required. Deploy Core 0.6.10 and re-test after the Laravel rollout settles; simultaneous upstream
503responses still require production runtime diagnosis.