Improvement

The official Improvement integration lets you provide feedback on your bot’s responses in the emulator. Your bot can then use that feedback when generating future responses.

Setup

The Improvement integration comes pre-installed on every bot. If you need to install it manually:

In Botpress Studio, select Explore Hub in the upper-right corner.
Search for the Improvement integration, then select Install Integration.
In the Configuration menu, select Save Configuration.

Suggesting improvements

You can suggest improvements when chatting with your bot in the emulator:

Select Improve Response on any message:

Improve response

Enter your feedback and select Re-generate., This will show you a preview of an updated response based on your feedback:

Feedback

The Code section shows you exactly how your bot’s execution will be different after the feedback is applied.

If you’re happy with the updated response, select Save as learning.

Managing feedback

Any time you save feedback, it gets added to the Feedback Table as a row:

Feedback

Your bot reads directly from this table when applying your feedback. This means you can modify the table’s rows and columns to modify how feedback is applied.

You can also delete a specific piece of feedback by deleting its associated table row.

Cards

Delete Feedback Entry for Iteration

Deletes feedback on an iteration for active learning.

inputobject
outputobject

Find Feedback

Finds relevant feedback entries for active learning.

inputobject
outputobject

Get Iteration

Fetches a single iteration by its ID.

inputobject
outputobject

List Conversation Iterations

Finds LLMz iterations for active learning.

inputobject
outputobject

List Feedback Entries

List all the feedback entries, optionally filtered by URI.

inputobject
outputobject

Post Install

This action is called after the integration is installed

inputobject
outputobject

Create or Update Feedback Entry for Iteration

Provides feedback on an iteration for active learning.

inputobject
outputobject

Track Iterations

Tracks LLMz iterations for active learning. This action should be called in an 'after-llmz-execution' hook.

inputobject
outputobject

Triggers