Skip to content

A2A Multi-Agent Fact Checker

Project Overview

The A2A Multi-Agent Fact Checker is a sophisticated demonstration of collaborative multi-agent systems built with Google's Agent2Agent SDK (A2A). This project showcases how specialized AI agents can work together under orchestration to verify facts and claims through a combination of external research and internal reasoning.

What Makes This Special?

  • Zero Configuration: Run with a single Docker command
  • Local LLM Inference: Uses Docker Model Runner (no API keys required)
  • Real-time Web Search: Integrates DuckDuckGo via Model Context Protocol (MCP)
  • Orchestrated Workflow: Demonstrates true agent collaboration
  • Modular Architecture: Each agent has distinct roles and capabilities

Architecture Overview

fact chcker

Agent Role Agent Tools Used Role Description
Auditor ❌ None None Coordinates the entire fact-checking workflow and delivers the final answer.
Critic ✅ DuckDuckGo via MCP DuckDuckGo via MCP Gathers evidence to support or refute the claim.
Reviser ❌ None None Refines and finalizes the answer without external input.

Getting Started

Before running the project, ensure you have:

  • Docker Desktop 4.43.0+ or Docker Engine installed
  • A laptop or workstation with a GPU (e.g., MacBook) for running models locally
  • Compose 2.38.1+ (if using Docker Engine on Linux)
  • GPU drivers properly installed for your system

Run the project

cd compose-for-agents/a2a
docker compose -f compose.yaml -f compose.offload.yaml up --build

By now, you should be able to see the containers as shown in the Docker Dashboard.

a2a-terminal

Open http://localhost:8080/dev-ui/?app=AgentKit in your browser to access the AgentKit UI, where you can interact with the agents.

a2a-ux

Prompt 1

Example 1: Scientific Fact Check

Input: "How far is moon from the earth?"

Workflow:

Auditor receives the question and plans the verification strategy Critic searches DuckDuckGo for current scientific theories and evidence Reviser analyzes the gathered evidence and forms a reasoned conclusion Auditor presents the final verdict with supporting evidence

Prompt 2

Example 2: Current Events Check

Input: "What is the current status of renewable energy adoption globally?"

Workflow:

Auditor identifies the need for recent data Critic searches for latest statistics and reports Reviser synthesizes multiple sources into coherent trends Auditor provides up-to-date analysis with data backing