We’ve introduced a new configuration option in Collect Data Block that gives you more precise control over how custom Escape conditions work.

What’s new?

You can now decide whether a custom Escape condition should consider messages sent before entering the Collect Data block, or only messages sent within the current block.

This setting is independent of “Scan historical messages for data”, allowing you to control separately:

• how data is extracted, and

• how Escape conditions are evaluated.

Why does this matter?

This change helps avoid unintended behavior in more complex flows, especially when chaining multiple Collet Data blocks.

What stays the same?

• Built-in agent request detection (“I want to talk to a human”) is unchanged and still works based on the user’s latest message.

• The new setting applies only to custom Escape conditions in the Collect Data block.

• The change does not affect existing processes. We made sure that the scanning is enabled for them which corresponds to the behavior before the change.

• Other block types (e.g. Options block) are not affected since they currently don’t support custom Escape conditions.

How to use it

When configuring a custom Escape condition, you’ll see a new check-box that lets you choose whether the Escape logic should:

• include messages from before entering the block, or

• only consider messages sent while the block is active.