Skip to main content

HexaEight Agent - Python library for AI agent coordination, PubSub messaging, and task management

Project description

HexaEight Agent - Python Library

A Python wrapper for HexaEight Agent that enables secure multi-agent communication and coordination for AI systems.

What is HexaEight Agent?

HexaEight Agent provides identity management and secure communication for AI agents. This library allows agents built with any framework (CrewAI, LangChain, AutoGen, etc.) to:

  1. 🔐 Authenticate with unique identities
  2. 💬 Communicate securely with end-to-end encryption
  3. 📋 Coordinate tasks across multiple agents
  4. 🔒 Lock messages for exclusive processing
  5. Schedule messages and tasks
  6. 🌐 Bridge communication across agents
  7. 📤 Publish messages with direct, broadcast, and scheduled delivery
  8. Handle events with real-time processing and async iteration

Requirements

  • Python 3.8 or higher
  • .NET 8.0 Runtime
  • Access to Agentic IAM (HexaEight Token Server and PubSub Server)
  • HexaEight credentials (Client ID, Resource Name, Machine Token)

Installation

pip install hexaeight-agent

Quick Start

  1. Install HexaEight Licensed Machine Token on a host machine (Visit: https://store.hexaeight.com)

  2. Setup an Agentic IAM Server - Configure HexaEight Token Server + HexaEight PubSub Server

  3. Create an Application and get a Client ID

  4. Create Parent Agent using the licensed host machine:

    dotnet script create-identity-for-parent-agent.csx parent_config.json --no-cache
    
  5. Create Child Agents using the parent agent configuration:

    dotnet script create-identity-for-child-agent.csx child_01 parent_config.json --no-cache
    
  6. Test Parent Agent from licensed host machine using demo:

    python3 hexaeight_demo.py parent_config.json parent 
    
  7. Test Child Agent from any machine using the configuration files:

    python3 hexaeight_demo.py config_agent02.json child
    
  8. Establish Secure Communication across agents by sending messages, locking messages, creating tasks, etc.

Note: A Licensed machine token is required for creating parent and child agents. The parent and child agents are created as JSON configuration files. Once created, these agents remain active forever, even after the machine token expires.

Agent Types

Parent Agents: Can create and manage tasks, tied to the licensed host machine, have full administrative capabilities, and can coordinate multiple child agents.

Child Agents: Can run on any machine, created using parent agent configuration, can participate in tasks and communication, and inherit security from parent agent.

Usage Examples

Refer to hexaeight_demo.py for complete examples of all features.

License

Apache License 2.0 - See LICENSE file for details.

Support


HexaEight Agent - Enabling secure, scalable AI agent communication

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

hexaeight_agent-1.6.801.tar.gz (1.5 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

hexaeight_agent-1.6.801-py3-none-any.whl (1.5 MB view details)

Uploaded Python 3

File details

Details for the file hexaeight_agent-1.6.801.tar.gz.

File metadata

  • Download URL: hexaeight_agent-1.6.801.tar.gz
  • Upload date:
  • Size: 1.5 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.3

File hashes

Hashes for hexaeight_agent-1.6.801.tar.gz
Algorithm Hash digest
SHA256 3797a952b4ef5dbb9ec8123ba4d45fbf32bfcb11e2cf9c823d5ccc54fdef3583
MD5 9e6ee34199ba850d21271445fcb438bb
BLAKE2b-256 c4fa13945e8b49af83d52f25d73b6cb903f4b387e516b3d64ef5f425b6871390

See more details on using hashes here.

File details

Details for the file hexaeight_agent-1.6.801-py3-none-any.whl.

File metadata

File hashes

Hashes for hexaeight_agent-1.6.801-py3-none-any.whl
Algorithm Hash digest
SHA256 cec0949e790f24b043192b9025700618d4cc32823ccde1d5ba61926299d7a92c
MD5 245210f0499e9902f3dd9b6ccf408d4b
BLAKE2b-256 30814ed117e7afb9d8efe764e7236a074ce06be598c9931ffbe1ab0433e90a04

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page