bolna_create_disposition
Create a call extraction disposition linked to a Bolna agent, defining how call transcripts are evaluated using an LLM with subjective or objective response options.
Instructions
Create a new call extraction disposition and link it to a Bolna agent
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| name | Yes | Display name for the disposition | |
| model | No | LLM model to use for evaluation | gpt-4.1-mini |
| agent_id | Yes | UUID of the agent this disposition will be linked to | |
| category | No | Category grouping for the disposition (default: "General") | General |
| question | Yes | The prompt sent to the LLM to evaluate the transcript | |
| is_objective | No | Enable pre-defined value selection | |
| is_subjective | No | Enable free-text response | |
| system_prompt | No | Optional system context for the evaluating LLM | |
| subjective_type | No | Format constraint for the free-text response | text |
| objective_options | No | Required when is_objective is true | |
| subjective_type_config | No | Required when subjective_type is 'regex' |