---
alias: generativeaiprompt-st
docsync: true
description: "This documentation outlines a table for configuring prompts within GenerativeAI actions"
---

# GenerativeAIPrompt

## Description

SmartTable to map the prompt prefix for a specific GenerativeAI Action and EntityType to use as context for the LLM

## Fields

The **GenerativeAIPrompt** table has the following fields:

| Field | Description | Key | Mandatory | Comments |
| --- | ----- | -- | -- | ---- |
| **EntityType** | The Entity Type name. | Yes | No | |
| **PromptId** | The prompt identifier. | Yes | Yes | |
| **AgentId** | The AI Agent identifier. | No | No | |
| **MCPServerIds** | A semi-colon list of MCP Server identifiers that will be pre-selected for this prompt. | No | No | |
| **Model** | The providers model to use. | Yes | No | |
| **ThinkingEnabled** | Defines whether extended thinking is enabled for the model. | No | No | |
| **Temperature** | The defined temperature for the model. Note: Must be a number between 0-1. | No | No | |
| **PromptPrefix** | The prompt prefix to give context to the LLM. | No | Yes | |
| **FirstPrompt** | An automatic initial prompt sent on the users behalf. | No | No | |

Table: GenerativeAIPrompt fields

## Precedence Keys

The search precedence rules are the following:

1. Model, EntityType, PromptId
2. EntityType, PromptId
3. PromptId