<aside> <img src="/icons/info-alternate_green.svg" alt="/icons/info-alternate_green.svg" width="40px" /> Tasks Axons are essential components in a Agent that handle the processing of data, whether it’s documents, raw data, or other forms of input. These axons allow you to classify, extract, compare, generate content, apply conditional logic to your data, to perform various transformations, such as generating documents, renaming files, merging files, performing other data adjustments or other kinds of processing.
</aside>
The Classifier Axon allows you to categorize documents or data based on labels that you define. It’s a powerful tool for organizing and sorting information automatically.
• Name: Define a unique name for each label. This will be used to classify documents or data under the appropriate label.
• Description: Provide a description of each label, clearly defining what types of documents or data fall under it.
The Extractor Axon allows you to extract specific pieces of information from your documents or data, making it easy to pull out the key details you need.
• Name: Specify a unique name for each datapoint. This name will be used to identify the extracted information.
• Description: Provide a brief description of the datapoint to clarify its purpose.
• Options: For certain datapoints, you can define specific values to extract by providing a list of predefined options.
This axon enables you to get information from a knowledge base and use it in the agent.
• Prompt: Specify how to use the input data to query the knowledge base and how to format the retrieved information.
• {{content}}: References the input data in your prompt.
• {{knowledge_base_result}}: Will be replaced with the information retrieved from the knowledge base.
The Decision Tree Axon enables you to create conditional logic for your templates. Based on the conditions you define, the axon can insert specific text or trigger actions as it processes the data.