Implementation:CrewAIInc CrewAI MySQL Search Tool

Knowledge Sources	CrewAI
Domains	Tools, RAG, Database
Last Updated	2026-02-11 00:00 GMT

Overview

MySQLSearchTool enables semantic search within MySQL database tables using RAG technology for natural language queries.

Description

The MySQLSearchTool extends RagTool with MySQL-specific functionality, transforming traditional SQL table access into semantic search. It requires a db_uri field for database connection. During initialization, it accepts a table_name, constructs a SELECT * query for that table, and adds the result to the RAG knowledge base with DataType.MYSQL and the database URI as metadata. The tool description is updated to reference the specific table name. The overridden add() method wraps table names in SELECT * FROM queries before delegating to the parent. The _run() method accepts semantic search queries with optional similarity_threshold and limit parameters, delegating to the parent RagTool's semantic search capabilities for natural language matching against the indexed table data.

Usage

Use this tool when a CrewAI agent needs to query MySQL database content using natural language rather than SQL syntax, making database exploration more intuitive and accessible for AI agents.

Code Reference

Source Location

• Repository: CrewAI
• File: lib/crewai-tools/src/crewai_tools/tools/mysql_search_tool/mysql_search_tool.py
• Lines: 1-46

Signature

class MySQLSearchTool(RagTool):
    name: str = "Search a database's table content"
    description: str = "A tool that can be used to semantic search a query from a database table's content."
    args_schema: type[BaseModel] = MySQLSearchToolSchema
    db_uri: str = Field(..., description="Mandatory database URI")

    def __init__(self, table_name: str, **kwargs): ...

    def add(self, table_name: str, **kwargs: Any) -> None: ...

    def _run(
        self,
        search_query: str,
        similarity_threshold: float | None = None,
        limit: int | None = None,
        **kwargs: Any,
    ) -> Any: ...

Import

from crewai_tools import MySQLSearchTool

I/O Contract

Inputs

Name	Type	Required	Description
db_uri	str	Yes	MySQL database connection URI
table_name	str	Yes	Name of the MySQL table to index for semantic search
search_query	str	Yes	Natural language search query to run against the table content
similarity_threshold	float or None	No	Minimum similarity score for returned results
limit	int or None	No	Maximum number of results to return

Outputs

Name	Type	Description
_run() returns	Any	Semantically matched content from the MySQL table

Usage Examples

Basic Usage

from crewai_tools import MySQLSearchTool

tool = MySQLSearchTool(
    table_name="products",
    db_uri="mysql://user:password@localhost:3306/mydb",
)
result = tool._run(search_query="high performance laptop computers")

Related Pages

• Principle:CrewAIInc_CrewAI_Built_In_Tool_Selection