v0.73.0
July 25, 2026
Work assigned to a name that doesn't exist yet no longer disappears in silence. You can still type a person's name as the assignee before you've even created their user — that's normal while you sketch a process — but now the system tells you, instead of leaving that work with no alerts and no reminders.
-
Feature
An assignee who doesn't exist is now visible
Assigning to a name, an email or a placeholder keeps working: it's a legitimate stage of designing a process, when the organisation isn't configured yet. What's new is that when that name matches no active user, the item flags it with a warning and the list marks it. Previously that work simply never produced notifications or reminders, and there was no way to notice. The warning disappears by itself the day you create the real user.
-
Feature
A view to reconcile placeholders
A new view groups every unrecognised assignee and shows where each one appears: items, tasks, phases, flow settings. When you get around to creating the real users, you reconcile them all at once instead of discovering them one at a time as you work.
-
Integration
A dedicated event for integrators
If you connect FlowSharp to other systems, you now receive an event when an assignment matches no user, with references to the record and the task in the payload. The event fires once, when the assignment is created or changed: it never turns into a repeated notification on every check.
-
Improvement
Impossible settings rejected on save
From the settings screen it was still possible to save an SLA of zero, negative or with decimals, and a flow like that was already overdue on every record: it's now rejected, as it already was from the designer, while SLAs computed from a variable remain valid. Naming a non-existent role among those who can start a flow or add tasks didn't widen the permission, it silently closed it to everyone: now it's flagged. The same start event no longer duplicates in the configuration. And the external entity link, which reported "saved" from settings without saving anything, is now actually applied.
-
Improvement
Duplicate records spotted even by eye
A record code copied from an external system or typed with an East Asian input method can contain full-width characters: to the eye it's the same text, to the system they were two different entities, and two records were opened for the same thing. They're now recognised as the same entity, using the same rule already applied to flow names. Hyphens and internal spaces stay distinctive, as they should.
-
Integration
More precise API errors on duplicate records
Trying to open a record for an entity that already has an active one now responds with a dedicated code, distinct from the one for a flow that can't be started. Previously the two situations were indistinguishable and an automation couldn't tell whether to resume the existing record or publish the flow. If you rely on that code, update your logic.
-
Security
Checks extended to entity links and role names
The external entity link is now checked when you create a flow too, not only when you edit it. Role names are stripped of markup, as flow names already were.
-
Improvement
User guide updated
The guide now covers operational lists and per-flow views, item notes and AI summaries, operating procedures, multichannel alerting, API surfaces for data and KPIs, flow settings and the field registry, configurable KPIs, workspace and live collaboration, the automation plan and Saving Challenge from the map.