Mindsdb (Official Sample)

MindsDB Enterprise Data Agent

Powered by Gemini 2.5 flash + MindsDB. This sample uses A2A to connect, query and analyze data across hundreds of federated data sources including databases, data lakes, and SaaS applications.

The agent takes natural language queries from users and translates them into appropriate SQL queries for MindsDB, handling data federation across multiple sources. It can:

• Query data from various sources including databases, data lakes, and SaaS applications
• Perform analytics across federated data sources
• Handle natural language questions about your data
• Return structured results from multiple data sources

Prerequisites

• Python 3.9 or higher
• MindsDB account and API credentials (https://mdb.ai)
• Create a MindsDB Mind (an AI model that can query data from a database), by default we use the demo one: Sales_Data_Expert_Demo_Mind

Environment Variables

In mdb.ai, once you create a Mind (an AI model that can query data from a database), you can use it in the agent.

Create a .env file in the project directory with the following variables:

MINDS_API_KEY=your_mindsdb_api_key
MIND_NAME=your_mindsdb_model_name

• MINDS_API_KEY: Your MindsDB API key (required)
• MIND_NAME: The name of the MindsDB Mind to use (required)

Running the Sample

1. Navigate to the samples directory:

   cd samples/python/agents/mindsdb
   

2. Run the agent:

   uv run .
   

3. In a separate terminal, run the A2A client:

   # Connect to the agent (specify the agent URL with correct port)
   cd samples/python/hosts/cli
   uv run . --agent http://localhost:10006
   
   # If you changed the port when starting the agent, use that port instead
   # uv run . --agent http://localhost:YOUR_PORT
   

4. Ask a question to the agent about your data.

Example Queries

You can ask questions like:

• "What percentage of prospects are executives?"
• "What is the distribution of companies by size?"

The agent will handle the complexity of joining and analyzing data across different sources.

Disclaimer

Important: The sample code provided is for demonstration purposes and illustrates the mechanics of the Agent-to-Agent (A2A) protocol. When building production applications, it is critical to treat any agent operating outside of your direct control as a potentially untrusted entity.

All data received from an external agent—including but not limited to its AgentCard, messages, artifacts, and task statuses—should be handled as untrusted input. For example, a malicious agent could provide an AgentCard containing crafted data in its fields (e.g., description, name, skills.description). If this data is used without sanitization to construct prompts for a Large Language Model (LLM), it could expose your application to prompt injection attacks. Failure to properly validate and sanitize this data before use can introduce security vulnerabilities into your application.

Developers are responsible for implementing appropriate security measures, such as input validation and secure handling of credentials to protect their systems and users.