Deep Research Agent for Sentient Chat

A sophisticated multi-agent research system that delivers comprehensive, well-structured research reports on complex topics. This agent orchestrates specialized sub-agents to break down research tasks, gather information, analyze content, and synthesize findings into detailed reports.

Features

• Multi-agent Architecture: Coordinated system of specialized agents working together
• Comprehensive Research Planning: Breaks down queries into specific research tasks
• Deep Information Retrieval: Simulates advanced information retrieval with reranking
• Insightful Content Analysis: Extracts key insights and patterns from retrieved information
• Professional Report Generation: Synthesizes findings into well-structured research reports
• Persistent Memory: Maintains research context across interactions
• Streaming Progress Updates: Provides real-time updates throughout the research process

Requirements

• Python 3.8+
• Fireworks.ai API key
• Jina AI API key (optional, for semantic reranking)

Installation

Option 1: Install from source

git clone https://github.com/yourusername/research-agent.git
cd research-agent
pip install -e .

Option 2: Install using pip

pip install git+https://github.com/yourusername/research-agent.git

AWS EC2 Deployment

To deploy the Research Agent on AWS EC2:

1. Launch an EC2 instance:

   • Recommended: t3.medium or larger
   • Ubuntu Server 20.04 LTS
   • At least 20GB storage

2. Connect to your instance:

   ssh -i your-key.pem ubuntu@your-instance-public-dns

3. Install dependencies:

   sudo apt update
   sudo apt install -y python3-pip python3-venv git

4. Clone and install the agent:

   git clone https://github.com/yourusername/research-agent.git
   cd research-agent
   python3 -m venv venv
   source venv/bin/activate
   pip install -e .

5. Set up environment variables:

   echo 'export FIREWORKS_API_KEY="your-fireworks-api-key"' >> ~/.bashrc
   echo 'export JINA_API_KEY="your-jina-api-key"' >> ~/.bashrc
   source ~/.bashrc

6. Run the agent with screen (keeps it running after you disconnect):

   screen -S research-agent
   cd ~/research-agent
   source venv/bin/activate
   python main.py --port 8000

   Press Ctrl+A then D to detach from the screen.

   To reattach later:

   screen -r research-agent

7. Set up a systemd service (alternative to screen):

   sudo nano /etc/systemd/system/research-agent.service

   Add the following content:

   [Unit]
   Description=Research Agent for Sentient Chat
   After=network.target
   
   [Service]
   User=ubuntu
   WorkingDirectory=/home/ubuntu/research-agent
   ExecStart=/home/ubuntu/research-agent/venv/bin/python main.py --port 8000
   Restart=always
   Environment="FIREWORKS_API_KEY=your-fireworks-api-key"
   Environment="JINA_API_KEY=your-jina-api-key"
   
   [Install]
   WantedBy=multi-user.target
   

   Enable and start the service:

   sudo systemctl enable research-agent
   sudo systemctl start research-agent

   Check status:

   sudo systemctl status research-agent

8. Set up NGINX as a reverse proxy (optional):

   sudo apt install -y nginx
   sudo nano /etc/nginx/sites-available/research-agent

   Add the following content:

   server {
       listen 80;
       server_name your-domain.com;
   
       location / {
           proxy_pass http://localhost:8000;
           proxy_http_version 1.1;
           proxy_set_header Upgrade $http_upgrade;
           proxy_set_header Connection 'upgrade';
           proxy_set_header Host $host;
           proxy_cache_bypass $http_upgrade;
       }
   }
   

   Enable the site and restart NGINX:

   sudo ln -s /etc/nginx/sites-available/research-agent /etc/nginx/sites-enabled/
   sudo nginx -t
   sudo systemctl restart nginx

9. Set up SSL with Certbot (optional):

   sudo apt install -y certbot python3-certbot-nginx
   sudo certbot --nginx -d your-domain.com

Usage

Running the agent directly

# Set your API keys
export FIREWORKS_API_KEY="your-fireworks-api-key"
export JINA_API_KEY="your-jina-api-key"  # Optional

# Run the agent
python main.py --port 8000

Using the agent in Sentient Chat

Once deployed, the agent can be accessed through the Sentient Chat interface by configuring the agent endpoint URL.

Configuration Options

The agent accepts the following command-line arguments:

• --host: Host to bind the server to (default: 0.0.0.0)
• --port: Port to bind the server to (default: 8000)
• --debug: Enable debug mode (default: False)
• --fireworks-api-key: Fireworks API key (alternative to using environment variable)
• --jina-api-key: Jina API key (alternative to using environment variable)

Architecture

The Research Agent is built using a multi-agent architecture with specialized components:

1. Research Coordinator: Manages the overall research process
2. Research Planner: Analyzes queries and creates research plans
3. Information Retriever: Gathers information from various sources
4. Content Analyzer: Evaluates and extracts insights from information
5. Report Generator: Synthesizes findings into comprehensive reports
6. Research Memory: Maintains state and tracks research progress

License

This project is licensed under the MIT License - see the LICENSE file for details.