AgentRouterInstall

Summary

Send a chat message to the AgentRouter API using a specified model and optional system prompt, receiving the model's text response.

Inputs

• model (required) — Select the chat model by name or ID from a dynamically loaded list, or specify an ID via expression.
• system_prompt — Provide a system message to define the assistant's behavior in the chat session.
• message (required) — Enter the user message to send to the chat model.
• temperature — Set the sampling temperature to influence response randomness (default 0.9).
• additionalFields — Specify optional parameters including frequency penalty, max tokens, presence penalty, and top_p to customize generation behavior.

Output shape

a JSON object with a 'response' field containing the model's generated reply text

Outputs one item per input with the model's trimmed response text. If errors occur per item, they are returned in the output JSON under 'error' field if 'Continue On Fail' is enabled.