> ## Documentation Index
> Fetch the complete documentation index at: https://pathlit.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Router Node

> Route workflow execution using AI-powered decision making

## Overview

The Router Node uses AI to intelligently route workflow execution to different branches. This node enables you to:

* Make routing decisions based on natural language analysis
* Define multiple output branches with descriptions
* Let AI determine the best path based on input content
* Create dynamic workflows that adapt to different scenarios
* Replace complex conditional logic with intelligent routing

## Configuration Parameters

### Node Configuration

* **Prompt**:

  The input or question that the router should analyze to make a routing decision.

  > *What language is this text written in? Text: \{input}*

  > *What category does this support ticket belong to? Ticket: \{ticket\_content}*

  > *Based on this customer feedback, what action should we take?*

* **Options**:

  Define the possible routing branches. Each option has:

  * **Name**: The branch identifier
  * **Description**: Explanation of when this route should be taken

<AccordionGroup>
  <Accordion title="Advanced Settings">
    <ul>
      <li>
        <strong>Model</strong>: Select the language model for decision making
      </li>

      <li>
        <strong>System Prompt</strong>: Customize how the AI makes routing decisions
      </li>

      <li>
        <strong>Temperature</strong>: Controls randomness (0.0-2.0, default 0.1 for consistent routing)
      </li>

      <li>
        <strong>Top P</strong>: Controls diversity (0.0-1.0, default 0.01 for deterministic routing)
      </li>
    </ul>
  </Accordion>
</AccordionGroup>

## Expected Inputs and Outputs

* **Inputs**:

  * The node accepts text input that can be referenced in the prompt

* **Outputs**:

  * **reasoning**: AI's explanation for the chosen route
  * **selected\_option**: The name of the selected branch
  * Dynamic output ports for each defined option (only the selected branch activates)

## Use Case Examples

1. **Language-Based Routing**:

   Automatically route customer messages to language-specific support teams based on detected language.

2. **Priority Triage**:

   Analyze support tickets and route them to appropriate departments or priority queues based on urgency and topic.

3. **Content Classification**:

   Classify incoming documents and route them to different processing pipelines based on their type or category.

## Error Handling and Troubleshooting

* **Unclear Routing**:

  If the AI cannot confidently choose a route, provide clearer option descriptions or more specific prompts.

* **Model Selection**:

  Use models with strong reasoning capabilities for complex routing decisions. Lower temperatures ensure consistent routing.

* **Option Descriptions**:

  Write clear, distinct descriptions for each option to help the AI make accurate routing decisions.

If you encounter any issues not covered in this documentation, please reach out to our [support team](mailto:dev@pathlit.ai) for assistance.

## Relevant Nodes

<CardGroup>
  <Card title="Conditional Node" href="/docs/nodes/conditional">
    Route based on rules and comparisons
  </Card>

  <Card title="LLM Node" href="/docs/nodes/llm">
    Process content before routing
  </Card>

  <Card title="Human Input Node" href="/docs/nodes/human-input">
    Let humans make routing decisions
  </Card>
</CardGroup>