Super Agent icon

Super Agent

Tool-calling AI agent that responds to user queries using connected language models and tools

Super Agent

AI-generated

Overview

The Super Agent node is an AI-powered agent designed to execute complex tasks by leveraging connected language models, memory modules, and tools. It dynamically routes prompts, maintains conversation context, and orchestrates tool usage to provide specialized and comprehensive responses. This node is beneficial in scenarios where multi-step AI interactions are needed, such as customer support automation, intelligent task execution, or conversational AI workflows that require tool integration and memory retention.

Use Case Examples

  1. Using the node to automatically extract user prompts from previous nodes and respond using a connected language model with access to external tools.
  2. Manually defining prompts and system messages to customize the AI assistant's behavior for specific tasks.
  3. Enabling memory connections to maintain conversation context across multiple interactions.
  4. Utilizing output parsers to format or process the AI's response before outputting.

Properties

Name Meaning
agentInfo Information notice about the node's capabilities and purpose.
promptType Determines whether the prompt is taken automatically from the previous node or manually defined below.
text The manually defined prompt text to send to the AI, required if promptType is 'define'.
systemMessageSource Selects whether to use the system message defined in this node or to delegate system message routing to a connected Super Prompts node.
systemMessage The system message that defines the AI agent's role and behavior, used if systemMessageSource is 'internal'.
enableOutputParser Whether to use a connected Output Parser node to process the AI response before outputting.

Output

JSON

  • output - The AI agent's response after processing, including tool calls and memory context.
  • chatInput - The prompt text sent to the AI agent.

Dependencies

  • Requires connection to an AI language model node (mandatory).
  • Optional connections to memory nodes to maintain conversation context.
  • Optional connections to tool nodes to enable tool usage by the AI agent.
  • Optional connection to an output parser node if output parsing is enabled.

Troubleshooting

  • Error if no language model is connected: Ensure an AI language model node is connected to the 'Chat Model' input.
  • Tool execution errors: Verify that connected tools are correctly implemented and callable.
  • Model does not support function calling/tools: Use a language model that supports function calling, such as OpenAI's GPT-4 or GPT-3.5-turbo.
  • Output parser enabled but not connected: Connect an output parser node or disable the output parser option.
  • Memory save/load failures: Check that memory nodes implement the expected interfaces for saving and loading chat history.

Links

Discussion