This node combines an agent with a set of tools and prompts it with a user provided prompt and the conversation history in the input table.
The conversation table is expected to have at least two string columns that define previous conversation. It can be empty if this is the start of the conversation.
The agent always receives the full conversation table as context which can slow down agent execution for long conversations or even lead to execution failures if the context becomes too large for the underlying LLM. If you experience such issues, you can truncate the conversation table by only keeping the last few messages, or even use a large language model to create a summary of the conversation held so far.
If checked activates prints the output of LangChain's debug mode to the console.
Column that specifies the sender role of the messages. The usual values are Human and AI.
Column containing the message contents that have been sent to and from the model.
The (next) message to send to the agent.
The agent to prompt.
The tools the agent can use.
Table containing the conversation that was held so far with the agent. Has to contain at least two string columns. A content column containing the content of previous messages and a role column specifying who the message is from (values hould be either AI or Human).
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 AI Extension (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, follow @NodePit on Twitter or botsin.space/@nodepit on Mastodon.
Please note that this is only about NodePit. We do not provide general support for KNIME — please use the KNIME forums instead.