Skip to content
Nappai Documentation

Runnable Executor

This component in Nappai’s dashboard lets you run automated tasks. Think of it as a universal task runner: you give it a task, and it handles the details of running it and giving you the answer.

Relationship with Runnables

This component works with “runnables,” which are pre-built pieces of code designed to perform specific actions. You don’t need to understand how the runnable works internally; you just need to provide the necessary input and the component will handle the execution.

Inputs

  • Input: This is the data you want to give to the task. For example, if the task is to translate text, this would be the text you want translated. This input is required.
  • Agent Executor: This is the specific task you want to run. Nappai will provide a list of available tasks to choose from. This input is required.
  • Input Key (Advanced): This specifies how the “Input” data is identified within the task. Usually, you don’t need to change this; the default value (“input”) works in most cases.
  • Output Key (Advanced): This specifies how the result of the task is identified. Like “Input Key,” the default value (“output”) usually works fine.
  • Stream (Optional): This setting determines whether the task runs in a continuous stream (like receiving live updates) or just runs once and provides a final result. The default is “off” (False).

Outputs

The component produces a single output:

  • Text: This is the result of the task. It could be anything, depending on the task you chose: translated text, a calculated number, a summary, etc.

Usage Example

Let’s say you have a task that summarizes text.

  1. In the “Input” field, you would paste the text you want summarized.
  2. In the “Agent Executor” field, you would select the “Text Summarizer” task from the list provided by Nappai.
  3. The component will run the summarizer task using your input text.
  4. The “Text” output will contain the generated summary.

Templates

This component is used in the following Nappai templates: ‘Stock chat’, ‘Search Tools’, and ‘SqlAgent’. The exact configuration might vary depending on the template.

The “Runnable Executor” often works with other components that provide input or consume its output. For example:

  • Many components: The list of components this connects to is extensive and includes various data sources, AI models, and output methods. Refer to Nappai’s component library for a complete list and details.

Tips and Best Practices

  • Start by using the default values for “Input Key” and “Output Key.” Only change them if you know what you’re doing and the specific task requires it.
  • If your task involves processing large amounts of data, consider using the “Stream” option for more efficient processing.

Security Considerations

Ensure that the “runnable” you select is from a trusted source. Avoid running untrusted or unknown runnables, as they could potentially compromise your data or system. Always review the description and purpose of each runnable before using it.