Changelog
Follow up on the latest improvements and updates.
RSS
Picking the right template from a long list no longer means scrolling and guessing. The redesigned Choose template window now shows categories, your template list, and a live preview side by side, and adjusts to fit your screen size.
- Search and filter: search by template title, or filter by Header, Labels, Button Type, and Language
- Starred and Recently used: jump straight to the templates you send most
- Browse by category: narrow down by Marketing (General, Carousel) or Utility
- Live preview: check the full message and buttons before you send
Your team spends less time hunting for templates and is more likely to send the right one the first time.
You can now reply in a conversation your AI agent is handling, and the agent stays in it on standby. When you’re done, hand the conversation back to the agent.
- Reply to take over.The agent pauses and keeps reading the conversation, but it won’t send anything while you have control.
- Hand back with a note.Tell the agent anything it couldn’t see for itself, like a discount you just agreed with the customer via a phone call.
- See what the AI did.Every handover is recorded in the conversation. A new “AI handling” inbox lists every conversation an agent is in.
Before, the API couldn’t see agent availability. Changing a role meant deleting and recreating the staff member, and renaming a team meant rebuilding it. With the Platform API, you can now:
- Check availability: see if an agent is active or away, and when that changed
- Set availability: set an agent to active or away, and routing respects it immediately
- List staff and teams: pull staff with availability included, or one team’s members
- Update staff: change role, name and teams in one request, with history untouched
- Rename teams: change a team’s name without rebuilding it
improved
Flow Builder
Find flow variables faster with one search and clear step labels
Variables with the same name from different nodes were easy to mix up, and finding the right one took too long. The rebuilt variable picker in Advanced Flow Builder fixes both:
- One search box: covers the trigger, every node, Contact properties and Common properties together
- Search by step: type a node name, like “Customer sends you a message”, to see its variables
- Source labels everywhere: each variable shows its step, such as Customer sends you a message › Message content, in message text, conditions, filters and on the canvas
- Labels that stay current: renaming a node updates its labels straight away
Existing flows switch to the new design automatically and keep running exactly as before.
improved
Flow Builder
Logs now show which workflow removed a contact from a flow
When the
Remove contact from workflows
action ends a contact’s enrollment, the Remarks column in that flow’s Logs now reads “Contact removed by action node in <workflow name>“. In Advanced Flow Builder, the enrollment’s activity timeline names the source workflow too. Manual removals still show “You have canceled the flow”, so you can tell automated and manual exits apart at a glance.If a purchase flow ends your nurture sequence, you can see why a contact left in seconds, without opening every flow to check.
Your AI agent can answer from a new source within minutes, but the Knowledge Base used to show Training for up to an hour. Removed sources also disappeared from the list right away, even though the agent kept using them for a while.
Each source now shows a clear status:
- Adding / Ready: see exactly when your agent can use a source
- Ready with a warning icon: article generation stopped, so press Retry under Generated articles
- Couldn’t add: tells you whether to re-export the file or the problem is on our side
- Removing: stays visible until removal finishes, usually within 10 minutes
- One Remove action: replaces Unassign, Unlink and Bulk Unassign, and sources stay in your Library
A line above the list shows how many of the sources you just added are ready, so you know when to start testing.
Your customers already tell you what's working and what isn't in every chat, email, and call. Until now, turning those conversations into insights meant reading them manually or relying on surveys that only some customers answer.
CX Intelligence reads every customer conversation in your workspace, groups it by topic, and scores satisfaction the same way every time:
- Topics:See conversations grouped into themes and subtopics, with trends over time and the risks or opportunities behind them.
- CSAT:Get an AI satisfaction score from 0 to 5 for every conversation, based on speed, resolution and sentiment, plus tips for your team. No survey needed.
- Ask:Ask questions in your own words and get an answer drawn from all your conversations, or request a custom dashboard.
🧪 Want to see how it works? Explore CX Intelligent with sample data to discover the insights it can uncover, without affecting your real customer data.
If your business runs several brands, regions or business units, you can now give each one its own workspace and manage them all from one login.
With multi-workspace, you can:
- Keep data fully separate:Each workspace has its own contacts, conversations, channels, automations, and analytics, and none of it is visible in other workspaces.
- Give each team its own setup:Assign separate admins, roles and seats in every workspace.
- Move between workspaces in seconds: Switch from the sidebar or the workspace picker after you sign in, on web and mobile.
Previously, AI Agent Analytics showed one AI Agent at a time, so it was hard to see how much AI was handling across your workspace or how its reply speed compared with your team's.
The Analytics tab in SleekFlow AI is now one dashboard for all your AI Agents, with a date range and comparison period at the top.
- Active contacts who interacted with AI: see how many of your active contacts AI was involved with
- Conversation volume and reply time: track conversation volume and first reply time for AI and staff over time
- Exit conditions: see why AI sessions ended and how often AI handed over to staff
- Export analytics data: Download the data displayed on your dashboard as a single CSV file.
improved
AgentFlow
Test and improve AI agent responses more efficiently
Until now, generated test questions ran automatically and consumed AI credits before you could review them. You also had to open each test case individually to see the AI agent's response.
With the improved AgentFlow testing experience, you can:
- Review questions before running tests: Check generated, imported, and manually added questions before you click Run, giving you more control over AI credit usage.
- Review test results faster: See a preview of each AI response and its "Good" or "Poor" rating directly in the batch testing table, without opening every case.
- Understand how AI responds: Use the "How it answered" option in chat testing to see which knowledge sources and steps your AI agent used to generate a response.
- Review side-by-side: Review individual test cases in a side panel without losing sight of the batch testing table.
Load More
→