Anthropic - Parse text action

The Parse text action extracts specific information from your input text in Anthropic, based on guidelines you define. The extracted information is available as datapills.

Input

Input fieldDescription
ModelSelect the Anthropic model you plan to use. For example, Claude 3 Haiku.
Source textEnter the text you plan to parse.
Fields to identifyEnter the fields you plan to identify from the text, with a description for each field. Required fields take effect only at the top level. Nested fields are always optional.
Max tokensEnter the maximum number of tokens to generate before stopping. Defaults to 4096 if left blank.

Output

Output fields dynamically generate based on the fields you define in the Fields to identify field.

Last updated: