Chat
The Chat block starts a live chat session between the user and a large language model within a MindStudio workflow.
Start an interactive chat session with an LLM
The Chat block starts a live chat session between the user and a large language model within a MindStudio workflow. It accepts configuration options such as a system intro message — an optional welcome prompt shown at the start of the session — and conversation starters, which are pre-crafted messages that help users initiate the conversation. The block can also accept a model override for advanced use cases where a specific model configuration is required.
By default, a Chat block ends the workflow once the session concludes. For workflows that need to continue after the chat, two transition modes are available. Controlled mode displays an explicit button (labeled "Continue" by default, though the label is configurable) that moves the user to a specified next block. Dynamic mode defines a set of transition cases, each with a condition description, that the LLM evaluates against incoming user messages — if a message matches a condition, the workflow transitions to the designated block instead of generating a chat response.
When a transition occurs, the block can save the full chat history to a named variable and the user's last message to a separate named variable, making both available to downstream blocks. This makes the Chat block suitable for conversational interfaces, support bots, guided intake flows, and any workflow where user input needs to be collected through open-ended dialogue before proceeding to another step.
What you can build
Real-world workflows powered by the Chat block.
Customer Support Chatbot
Deploy a chat session as a front-line support interface, with a dynamic transition that routes users to a contact form if their message signals a need to reach a human agent.
Guided Onboarding Assistant
Use a Chat block with a welcome intro message and conversation starters to walk new users through setup questions before transitioning to a configuration step.
Lead Qualification Flow
Collect open-ended information from a prospect through a chat session, then save the conversation history to a variable and pass it to a downstream data-processing block.
Interactive FAQ Bot
Present users with pre-crafted conversation starters covering common questions, allowing them to get answers through a natural chat interface without navigating static content.
Intake Form Replacement
Replace a structured form with a conversational Chat block that gathers user details, then transitions in controlled mode to a summary or confirmation step.
Troubleshooting Workflow Agent
Build a diagnostic agent where the LLM chats with the user to identify an issue, and dynamic transitions invoke specific resolution steps based on what the user describes.
Ready to add Chat to your workflow?
Get Started FreeCommon questions about Chat
What are the required parameters for the Chat block?
The only strictly required fields are a unique id and the type value 'chat'. All other fields — including systemIntroMessage, conversationStarters, modelOverride, and transition settings — are optional.
What does the Chat block return or output?
The Chat block does not define a structured output field. However, when a transition is configured, it can write the full chat history to a variable specified by historyVariableName and the user's last message to a variable specified by lastMessageVariableName, both of which are then accessible to subsequent blocks.
What is the difference between controlled and dynamic transition modes?
Controlled mode shows an explicit button (default label: "Continue") that the user clicks to move to a specified next block. Dynamic mode defines an array of transition cases, each with a condition description; the LLM evaluates each incoming user message against these conditions and transitions to the matching block instead of responding, if a match is found.
What kinds of workflows commonly use the Chat block?
The Chat block is used in any workflow requiring open-ended user dialogue, such as support bots, onboarding assistants, lead qualification flows, and troubleshooting agents. Most Chat blocks terminate the workflow at the end of the session; transition modes are used when the workflow needs to continue to another block after or during the chat.
Can I customize the welcome message and suggested prompts shown to users?
Yes. The systemIntroMessage field accepts an optional string displayed at the start of the session, and conversationStarters accepts an array of strings representing pre-crafted messages the user can select to begin the conversation.
Related capabilities
Start
Specifies how the workflow is invoked
User Input
Prompt the user for one or more pieces of data that will be saved as a variable. Can include text, multiple choice, images, or files.
Run Function
Execute a custom JavaScript or Python function
End
End the workflow. Optionally return a structured output (for API) or send notifications.
Jump
Transition to another workflow, maintaining the same scope
Menu
Present the user with a menu and transition to the next step based on their choice
Add Chat to your workflow
Build powerful AI workflows with drag-and-drop blocks. No coding required.