Tools
Bot Tools allow your bot to perform actions beyond answering questions, such as interacting with systems, responding to website behavior, or escalating conversations to human agents.
All Bot Tools are organized into four tabs at the top of the Tools page:
- Preset
- Browser Function
- API Tools
- MCP
You can navigate between these tabs to configure different types of bot capabilities. Each tab represents a distinct integration approach, and the explanation below follows the same order as the tabs shown in the interface.
Preset
Preset tab provides built-in tools that can be enabled to extend your bot's default behavior without additional configuration. These tools handle common automation and interaction needs.
In this section, there are 3 tools that you can activate, such as:
Notify Human Agent
Tool that allows the bot to notify human agents via email when human assistance is required. In addition to sending a notification, the bot also includes the reason why human intervention is needed, so agents understand the context before taking over the conversation.
You can define when the bot should trigger human escalation and what response the bot sends to the customer. Common trigger scenarios include:
- When the bot does not know how to answer a customer's question
- When the customer explicitly asks to speak with a human agent
At the same time, you can configure the response message that the bot sends to the customer when this tool is triggered, informing them that a human agent will take over the conversation.
Next, you can define who will receive the notification email. You can:
- Select one or more agents within your workspace, or
- Notify people outside the workspace by entering their email addresses
By configuring the trigger conditions, customer response, and notification recipients, you ensure that conversations requiring human attention are escalated clearly and efficiently.
Image Handling
Tool that allows the bot to process and respond to messages that include images.
When this tool is enabled, customers can send images such as screenshots, photos, or visual references as part of their conversation. The bot can then use the image content as additional context when generating its response.
This tool is useful in scenarios where visual information is important, for example:
- Customers reporting visual issues or errors
- Users sharing screenshots for clarification
- Product-related questions that involve images
Auto-Labeling
Tools that allows the bot to automatically apply labels to conversations based on the detected intent or context.
When this tool is enabled, the bot analyzes incoming messages and assigns relevant labels without manual intervention. These labels can then be used for conversation organization, reporting, and automation workflows.
Auto-Labeling is useful for:
- Categorizing conversations by topic or intent
- Improving reporting and analytics accuracy
- Supporting automation rules that rely on labels
Configure Auto-Labeling
In this section, you can enable Auto-Labeling and define how the bot assigns labels based on conversation context.
Auto-Labeling Result in Conversation
Once Auto-Labeling is active, labels are automatically applied during live conversations. Agents can immediately see the assigned labels and use them for follow-up actions or routing.
Browser Function
Browser function tab allows your bot to respond to interactions with the website.
This feature will only work if the widget you use is an embed code that is attached to your website/platform and at least you have technical coding or scripting skills because this feature allows your bot to read the code/script on the website/platform.
API Tools
API Tools tab allows your bot to respond to website interactions through APIs. This enables deeper integrations by connecting the bot to external systems via API calls.
This feature will only work if the widget you use is an embed code that is attached to your website/platform and at least you have technical coding or scripting skills, especially in communication between systems using the API.
To learn how to configure API Tools and connect your bot to external systems, please follow this guide.
MCP
MCP tab allows your bot to communicate with external tools using the Model Context Protocol. This enables advanced integrations by passing structured context between the bot and supported external tools.
This feature will only work if the widget you use is an embed code that is attached to your website/platform and at least you have technical coding or scripting skills, especially in communication between MCP systems.
For further tutorials you can access the official guide at the following link https://modelcontextprotocol.io/tutorials/building-mcp-with-llms
