Click the blue text to follow me for more AI insights~Paying tribute to the extraordinary, honoring the ordinary

MultiAgentPPT is an intelligent presentation generation system that integrates A2A (Ask-to-Answer) + MCP (Multi-agent Control Protocol) + ADK (Agent Development Kit) architecture, supporting multi-agent collaboration and streaming concurrency mechanisms.

https://github.com/johnson7788/MultiAgentPPT

MultiAgentPPT

A multi-agent system based on A2A + MCP + ADK, supporting streaming concurrent generation of high-quality (online editable) PPT content.

🧠 Project Overview

MultiAgentPPT utilizes a multi-agent architecture to automate the process from topic input to complete presentation generation, with the main steps including:

1. Outline Generation Agent generates an initial content outline based on user requirements.
2. Topic Splitting Agent breaks down the outline content into multiple topics.
3. Research Agent Working in Parallel multiple agents conduct in-depth research on each topic.
4. Summary Agent Outputs compiles the research results to generate PPT content, returning it to the frontend in real-time streaming.

Advantages

• Multi-Agent Collaboration improves the efficiency and accuracy of content generation through parallel work by multiple agents.
• Real-time Streaming Return supports streaming return of generated PPT content, enhancing user experience.
• High-Quality Content generates high-quality content outlines and presentations by combining external retrieval and agent collaboration.
• Scalability the system is designed flexibly, making it easy to expand with new agents and functional modules.

Interface Screenshots

The following are core functionality demonstrations of the MultiAgentPPT project:

1. Topic Input Interface

The user inputs the desired PPT topic content in the interface:

2. Streaming Outline Generation Process

The system returns the generated outline structure in real-time based on the input content:

3. Complete Outline Generation

The final system will display the complete outline for user confirmation:

4. Streaming PPT Content Generation

After confirming the outline, the system begins to stream generate the content for each slide and returns it to the frontend.

Concurrent Multi-Agent Collaboration Process

Project Structure

Quick Start

Backend Environment Configuration (Python)

1. Create and activate a Conda virtual environment:

conda create --name multiagent python=3.12
conda activate multiagent

2. Install dependencies:

cd backend
pip install -r requirements.txt

3. Set backend environment variables:

# Copy template configuration files for all modules
cd backend/simpleOutline && cp env_template .env
cd ../simplePPT && cp env_template .env
cd ../slide_outline && cp env_template .env
cd ../slide_agent && cp env_template .env

Start Backend Service

Frontend Database Setup

Database for storing user-generated PPT:1. Start PostgreSQL using Docker:

docker run --name postgresdb -p 5432:5432 -e POSTGRES_USER=postgres -e POSTGRES_PASSWORD=welcome -d postgres

2. Install dependencies and push database models:

pnpm install
pnpm db:push

3. Insert default user (if the frontend reports no user, please try manually inserting a user data record:

frontend/insert_one_user.sql)

<span><span><span>4. .env</span></span></span> Example configuration:

DATABASE_URL="postgresql://postgres:welcome@localhost:5432/presentation_ai"
A2A_AGENT_OUTLINE_URL="http://localhost:10001"
A2A_AGENT_SLIDES_URL="http://localhost:10011"

Frontend Installation and Running (Next.js)

1. Install dependencies:

cd frontend
npm install

2. Configure environment variables:

cp env_template .env

3.Run the frontend development server:

npm run dev

4.Open the browser and visit:http://localhost:3000 Lao Zhu Reads AI Honoring the extraordinary Honoring the ordinaryClick ‘Looking’ if you think you are the best