whois

by PalNilsson

A minimal example of an MCP agent and server that can be used to ask an AI about who someone is. It provides a way to query different AI models for information about individuals.

View on GitHub

Last updated: N/A

What is whois?

This is a minimal example of an MCP agent and server designed to query AI models (like OpenAI, Anthropic, Gemini, and Llama) for information about individuals. It allows users to ask 'who is' questions and receive responses generated by the AI models.

How to use whois?

Install dependencies using pip install -r requirements.txt. 2. Set the required environment variables for API keys (ANTHROPIC_API_KEY, OPENAI_API_KEY, GEMINI_API_KEY, LLAMA_API_URL). 3. Run the server using uvicorn server:app --reload. 4. Run the agent with a query and specify the AI model to use (e.g., python agent.py "Albert Einstein" openai).

Key features of whois

Supports multiple AI models (OpenAI, Anthropic, Gemini, Llama)
Simple server and agent architecture
Easy to install and use
Uses environment variables for secure API key management

Use cases of whois

Quickly retrieving information about individuals using AI
Comparing responses from different AI models
Building AI-powered search applications
Educational purposes for understanding MCP agents

FAQ from whois

What AI models are supported?

The server supports OpenAI, Anthropic, Gemini, and Llama models.

How do I set up the API keys?

You need to set the ANTHROPIC_API_KEY, OPENAI_API_KEY, GEMINI_API_KEY, and LLAMA_API_URL environment variables with your respective API keys.

How do I run the server?

Use the command uvicorn server:app --reload.

How do I query the agent?

Use the command python agent.py "[Name]" [model], replacing [Name] with the person's name and [model] with the AI model (e.g., openai, anthropic, gemini, llama).

What is an MCP agent?

MCP likely refers to a Master Control Program agent, which in this context acts as an interface to query different AI models