Implementation:CrewAIInc CrewAI AI Mind Tool

Knowledge Sources	CrewAI
Domains	Tools, Database_Query, MindsDB
Last Updated	2026-02-11 00:00 GMT

Overview

Concrete tool for natural language querying of databases through MindsDB's AI Minds platform provided by CrewAI.

Description

The AIMindTool class extends BaseTool to create a wrapper around MindsDB's Minds API. During initialization, it accepts datasource configurations (engine, connection_data, tables, description) and creates a Mind instance via the MindsDB client with randomly generated names using the crwai_mind_ and crwai_ds_ prefixes. Queries are executed through an OpenAI-compatible API endpoint at https://mdb.ai/, allowing natural language questions to be answered from structured data sources including PostgreSQL, MySQL, MariaDB, ClickHouse, Snowflake, and Google BigQuery. The tool requires the MINDS_API_KEY environment variable and the minds-sdk package dependency.

Usage

Use this tool when CrewAI agents need to query structured databases using natural language without writing SQL. Ideal for data analysis tasks where agents must extract insights from connected data sources.

Code Reference

Source Location

• Repository: CrewAI
• File: lib/crewai-tools/src/crewai_tools/tools/ai_mind_tool/ai_mind_tool.py
• Lines: 1-102

Signature

class AIMindTool(BaseTool):
    name: str = "AIMind Tool"
    description: str = "A wrapper around AI-Minds..."
    args_schema: type[BaseModel] = AIMindToolInputSchema
    api_key: str | None = None
    datasources: list[dict[str, Any]] = Field(default_factory=list)
    mind_name: str | None = None

    def __init__(self, api_key: str | None = None, **kwargs): ...
    def _run(self, query: str): ...

Import

from crewai_tools import AIMindTool

I/O Contract

Inputs

Name	Type	Required	Description
api_key	str or None	No	MindsDB API key; falls back to MINDS_API_KEY env var
datasources	list[dict[str, Any]]	No	List of datasource configs with engine, description, connection_data, and tables
query	str	Yes	Natural language question to ask the AI Mind (runtime _run parameter)

Outputs

Name	Type	Description
_run() returns	str	AI-generated answer to the natural language query from the connected data sources

Usage Examples

Basic Usage

from crewai_tools import AIMindTool

tool = AIMindTool(
    api_key="your-minds-api-key",
    datasources=[{
        "engine": "postgres",
        "description": "Sales database",
        "connection_data": {
            "host": "localhost",
            "port": 5432,
            "database": "sales_db",
            "user": "user",
            "password": "pass"
        },
        "tables": ["orders", "customers"]
    }]
)

Related Pages

• Principle:CrewAIInc_CrewAI_Built_In_Tool_Selection