Internet of Agents

A distributed agent system that enables multiple agents to run and communicate with each other over HTTP/HTTPS.

Overview

This system allows you to run multiple agents with unique IDs, each operating on different ports for bridge and API communication. The agents can communicate with each other and with external services through a registry.

Prerequisites

• Python 3.x
• Bash shell
• Internet connectivity
• Anthropic API key
• Let's Encrypt SSL certificates
• Virtual environment (venv)

Configuration

The system requires a configuration file at /etc/internet_of_agents.env with the following environment variables:

# Required environment variables
ANTHROPIC_API_KEY="your-api-key-here"
AGENT_ID_PREFIX="your-prefix"
DOMAIN_NAME="your-domain.com"

# Optional environment variables
NUM_AGENTS=1  # Defaults to 1 if not specified

Agent Configuration

The agent system can be configured through the following parameters in start_running_agents.sh:

# Starting port numbers for bridge and API
START_BRIDGE_PORT=6000
START_API_PORT=6001

Port Configuration

• Bridge ports start from START_BRIDGE_PORT and increment by 2
• API ports start from START_API_PORT and increment by 2
• For example, with NUM_AGENTS=3:
  • Bridge ports: 6000, 6002, 6004
  • API ports: 6001, 6003, 6005

Agent IDs

• Agent IDs follow different patterns based on the domain:
  • For nanda-registry.com domains: agentm{AGENT_ID_PREFIX}{INDEX}
  • For other domains: agents{AGENT_ID_PREFIX}{INDEX}
• Example with AGENT_ID_PREFIX=6 and NUM_AGENTS=3:
  • For nanda-registry.com: agentm60, agentm61, agentm62
  • For other domains: agents60, agents61, agents62

Usage

1. Set up the environment file:

   sudo nano /etc/internet_of_agents.env
   # Add the required environment variables

2. Ensure SSL certificates are in place:

   • Certificate: /etc/letsencrypt/live/${DOMAIN_NAME}/fullchain.pem
   • Private key: /etc/letsencrypt/live/${DOMAIN_NAME}/privkey.pem

3. Make the script executable:

   chmod +x start_running_agents.sh

4. Run the script:

   ./start_running_agents.sh

Monitoring and Management

Check Running Agents

ps aux | grep run_ui_agent_https

Stop All Agents

for pid in logs/*.pid; do kill $(cat $pid); done

Logs

• Agent logs are stored in the logs directory
• Each agent has its own log file: logs/agentm{ID}_logs.txt
• Process IDs are stored in: logs/agentm{ID}.pid

Network Configuration

The system automatically detects the server's IP address using:

1. AWS checkip service
2. ifconfig.me service
3. Falls back to localhost.com if both fail

Each agent is configured with:

• Public URL: http://{SERVER_IP}:{BRIDGE_PORT}
• API URL: https://{DOMAIN_NAME}:{API_PORT}

Registry

Agents are registered with a central registry at:

• Registry URL: https://chat.nanda-registry.com:6900

Security

• API communication is secured using SSL certificates from Let's Encrypt
• Environment variables are stored in a system-wide configuration file
• Each agent runs in its own process
• Virtual environment isolation

Troubleshooting

1. If agents fail to start:

   • Check if ports are available
   • Verify environment variables in /etc/internet_of_agents.env
   • Check SSL certificate paths
   • Check logs in the logs directory

2. If agents can't communicate:

   • Verify network connectivity
   • Check if registry is accessible
   • Ensure ports are not blocked by firewall
   • Verify SSL certificate validity

License

MIT License

Copyright (c) 2024 Internet of Agents

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.