Documentation
УКР
Release
Article MarkdownReport an error

Release 5.11.0

5.11.0 is a major step forward for teams that build processes on data, internal knowledge, and fast support. We redesigned the Custom Data workspace, added flexible organization for the instance knowledge base, opened a staff view in Client Portal, and let scenarios react to internal ticket notes — with less manual work and without leaking internal text to the customer.

This release is about making ConnectiveOne feel more modern every day: fewer screen hops, more clear actions where you already are, and automation where teams previously relied on chat agreements.

Highlights

  • Custom Data — a new workspace. Create tables, import with preview, Kanban board, semantic field types, table trash, and stable work with large tables.
  • Instance knowledge base — tags and quick actions. Colored labels, filters, multiple processes per document, and metadata edits from cards or list rows.
  • Client Portal for staff. Operators and supervisors open the same ticket link as the customer — with internal notes and support context, read-only.
  • Scenarios react to internal notes. A new internal event starts automation when an operator leaves an internal ticket note.
  • Email in Operator Line without unnecessary blocks. Forward a message as-is or send attachments only — without mandatory text in the message field.
  • Easier node connections in Scenario Builder. Connection handles on the canvas are larger — building complex branches is less fiddly.
  • Fast Line Pro — Ask AI for meaningful text splitting. From knowledge-base segment settings you can hand a file to Instance Agent for semantic chunking.

Custom Data

Custom Data gets a refreshed interface: from “new table” start to everyday work with thousands of rows. The goal is that integrators and analysts share one clear workspace — not a technical builder “only if you already know JSON”.

What’s new

  • Module start page — tables listed by groups; create a table or open records in one click.
  • “New table” screen — import from file, structure starter, blank editor, or Ask AI (when Instance Agent is available on the instance).
  • Import with preview — sample rows and validation errors before write; Excel/CSV and Google Sheets by link.
  • Clear-and-import mode — full table replace only after confirming how many rows will be deleted.
  • Guided table editor — fields, statistics, and presentation in one place; groups are created in the form.
  • Records view — filters, saved views, sorting, grouping, and stats under the table.
  • Kanban board — switch a saved view to a board; dragging a card updates the field value.
  • Semantic field types — email, phone, link, image, color, and more render clearly in the table and form.
  • System reference fields — operator, topic, tag, priority, skill group show a label, not only an ID.
  • Scenario tab — ready action snippets and where the table is already used in bots.
  • Row macros — run a scenario from a record when configured for the table.
  • Table trash — restore a deleted table from Trash in the table list (rows stay until you Delete permanently).
  • Large tables without a dead end — if the chosen sort isn’t ready for a fast open yet, the table still opens in a safe order; a banner explains what actually applied, and an administrator can start sort acceleration with a risk estimate.

How it changed

Before: Custom Data felt like a technical tool — manual structure setup, import without confidence about what would be written, separate screens for groups, and limited comfort at large volumes.

Now: you follow a clear path from a table idea to a working view — with import preview, board mode, saved views, and protection against accidental data loss on large tables.

Why it matters: directories, case journals, and scenario data can be maintained faster and safer — without fear of “breaking” a table during import or sorting.

Please note (Custom Data)

  • An instance administrator enables the new Custom Data UI (how to enable). Until then, the familiar interface remains.
  • Structural changes (field type, deleting a field) can change column data — the system warns before save.
  • Ask AI depends on Instance Agent settings on your instance.
  • The old AI Magic model generator is removed from the UI; use Ask AI via Instance Agent instead.

More: Custom Data analyst hub, create a table, how large tables work.

Constructor — instance knowledge base

Internal team instructions no longer get lost in a flat list: tags, convenient filters, and quick actions without opening the full editor every time.

Tags and filters

  • Tags opens the instance label catalog (name + color).
  • A document can have multiple tags; clicking a tag filters the catalog.
  • In the catalog: search by name and document count; View applies a list filter quickly.
  • Filters popover — type, process, access, pinned state, tags; active conditions show as chips.
  • Cards and List layouts expose the same metadata actions.

Inline metadata

Before: type, bot, and status were visible but editable only in the full document editor; a document linked to only one bot.

Now: type, tags, processes (multiple per document), pin-to-home, and access (public / restricted) change from a card or list row; clicking type, tag, or process also filters the catalog. The UI uses the term process throughout.

More: how to use the instance knowledge base, what is the instance knowledge base.

Client Portal — staff view

Support gets closer to the customer experience: an organization employee can open the same ticket link as the reporter — and see internal context that end users never get.

What’s new

  • A dedicated role permission — Client Portal (staff) — for read-only access.
  • On the ticket page, staff see the public thread together with internal notes and a context panel (tracker, Operator Line, and more — per instance settings).
  • In Support Center, staff can open a ticket list (Show ticket list) within operator permissions, with organization and client filters — without create or edit.
  • Users with this permission are not blocked from Constructor — unlike the end-user “Client Portal only” role.
  • Tracker links in the response use a trusted address from instance settings.

How it changed

Before: an employee opening a customer ticket link saw the same screen as the reporter — no internal notes and no separate staff access via roles.

Now: an administrator grants Client Portal (staff) together with Operator Panel and ticket permissions, and sets tracker and Operator Line URLs. Operators get one URL from customer emails and notifications — without workarounds and without exposing internal notes to regular portal users.

More: staff view in Client Portal, configure staff access.

Scenario Builder

Internal ticket note

Before: a scenario with the “ticket message” event reacted only to the public thread; there was no separate trigger for an internal note. Chat reminders could mistake an internal note for a public operator reply.

Now: in Start → Internal events you can choose Internal ticket note. The note text is never sent to the customer; the scenario gets a short preview. Reminders no longer confuse an internal note with a public reply.

Why it matters: you can automate escalation, webhooks, or internal steps exactly when the team leaves a service mark — without leaking the note to the customer.

More: configure internal events, internal events reference.

Easier node connection handles

Connection points on the canvas (node inputs/outputs, including keyboard buttons) are larger — easier to aim and wire complex scenarios with fewer misses.

Operator Line

Email: forward and attachments without mandatory text

Before: forwarding a message as-is or sending attachments only required text in the message field; with an empty field, send could silently do nothing.

Now: on the email channel you can forward without an extra comment and send attachment-only messages. A truly empty send (no forward context, no files) shows a clear warning instead of silence.

Why it matters: operators hand off correspondence faster without inventing filler text in the reply field.

More on forwarding: manage email recipients.

Fast Line Pro

Ask AI for meaningful knowledge-base chunking

In document segment settings, Ask AI appears when Instance Agent is available (and not read-only): you can pass uploaded files to the agent for semantic chunking — splitting by meaning, not only by character count. Ready segments are saved through the usual knowledge-base segment flow.

Manual segment size settings in Fast Line Pro stay as they are — Ask AI complements them when smarter splitting is needed.

More: Fast Line Pro hub, split a file with Ask AI.

Please note

  • Custom Data UI v2 and Instance Docs may need separate enablement by the instance administrator; check with the ConnectiveOne team or instance settings.
  • After an Engine upgrade for the instance knowledge base (tags and multiple processes), your administrator may need the standard database schema sync.
  • Do not mix Client Portal (staff) with the end-user reporter role on the same role unless needed — staff takes priority, but mixed roles complicate auditing.
  • This release also includes quality fixes from the 5.9.x–5.10.x line (Operator Line, Fast Line Pro, administration). If you upgrade from an intermediate version, also check the 5.9.3–5.10.0 notes.
Using your own AI agent?Install the open skill so your AI agent can work with current official ConnectiveOne documentation.Get the skillNeed support?Couldn’t find the answer or need help from the ConnectiveOne team? Create a request in Client Portal.Create a request