This node enables interactive, multi-turn conversations with an AI agent, combining a chat model with a set of tools and optional input data.
The agent is assembled from the provided chat model and tools, each defined as a KNIME workflow. Tools can include configurable parameters (e.g., string inputs, numeric settings, column selectors) and may optionally consume input data in the form of KNIME tables. While the agent does not access raw data directly, it is informed about the structure of available tables (i.e., column names and types). This allows the model to select and route data to tools during conversation.
Unlike the standard Agent Prompter node, which executes a single user prompt, this node supports multi-turn, interactive dialogue. The user can iteratively send prompts and receive responses, with the agent invoking tools as needed in each conversational turn. Tool outputs from earlier turns can be reused in later interactions, enabling rich, context-aware workflows.
This node is designed for real-time, interactive usage and does not produce a data output port. Instead, the conversation takes place directly within the KNIME view, where the agent’s responses and reasoning are shown incrementally as the dialogue progresses.
To ensure effective agent behavior, provide meaningful tool names and clear descriptions — including example use cases if applicable.
Instructions provided by the workflow builder that guide the agent's behavior. It typically defines the agent’s role, its tone, boundaries, and behavioral rules.This message is prioritized over the user message and should not contain any information that the user can inject in order to prevent prompt injection attacks.
The column of the tools table holding the tools the agent can use.
An optional 'AI' initial message to be shown to the user.
If checked, the tool calls and the tool call results will also be shown in the chat.
The maximum number of times the agent can repeat its steps to avoid getting stuck in an endless loop. If not provided, defaults to 25.
In debug mode, tool executions are displayed as meta nodes in the agent workflow and the meta node is kept in case of an error in the tool.
Prefix for the data message shown to the agent. You can use this to customize the instructions about the data repository.
You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.
To use this node in KNIME, install the extension KNIME Python Extension Development (Labs) from the below update site following our NodePit Product and Node Installation Guide:
A zipped version of the software site can be downloaded here.
Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.
Try NodePit Runner!Do you have feedback, questions, comments about NodePit, want to support this platform, or want your own nodes or workflows listed here as well? Do you think, the search results could be improved or something is missing? Then please get in touch! Alternatively, you can send us an email to mail@nodepit.com.
Please note that this is only about NodePit. We do not provide general support for KNIME — please use the KNIME forums instead.