Skip to main content

v1.14 — April 2026

New features​

Tracing level — Complete mode​

The Complete mode configuration now includes a duration choice (between 5 minutes and 4 hours): after this period, the level automatically switches back to Performance, limiting the footprint while still leaving time to analyze an incident. The setting is applied consistently across the interface. For details on the three modes (Complete, Standard, Performance), see Tracing modes and Workflow version management (adjusting the level on a deployed version).

Fixes and behavior improvements​

Workflows, nodes and APIs​

  • HTTP responses from the Code node: the configuration is more flexible — headers are no longer required; you can define a response with only the status and body, simplifying most scenarios.
  • Condition node in Code: removed a behavior that could, in some cases, interrupt the workflow execution; the path is reliable again.
  • Condition node and Switch node: in detailed logs mode, all useful information (metadata) is visible in the trace, for a complete view of the execution — see Tracing modes.
  • AI Agent node after a loop (ForEach node): the link between what you assemble in the editor and what runs works again as expected in this use case.
  • Workflow rerun: the already deployed version is proposed by default, avoiding handling mistakes and reassuring during a replay — see Replay a trace and Deploy a workflow.
  • APIs and incoming requests: when the content type is not specified, the platform applies a clear and documented behavior (text/plain), reducing ambiguities on the integration side — see API route trigger.

FTP and file processing​

  • Large files: fixed a case where the same file could be processed twice; the path is now unique and controlled, avoiding instability risks — see the FTP nodes and the File trigger.
  • FTP-related schedules (CRON): after a connection loss, schedules resume automatically as expected, without manual intervention — see Scheduler trigger.

Triggers and dictionaries​

  • SQS trigger with no associated workflow: the alert remains relevant and now gives you access to the useful detail to understand the situation and act quickly — see Manage triggers.
  • Dictionaries: entry duplication is stable again; deployment also supports larger value sets (large arrays), covering more business realities — see Deploy a trigger.

User interface​

  • Workflow dashboard: corrected display for more comfortable viewing — see Workflows.
  • Navigation: side menus offer a more polished experience (revised scrollbar behavior).