Skip to main content

AI-driven task automation system

Project description

GitHub - LicensePyPIDocumentationPython Version

Homepage

AgentForge

AgentForge is a low-code framework tailored for the rapid development, testing, and iteration of AI-powered autonomous agents and Cognitive Architectures. Compatible with a range of LLM models — currently supporting OpenAI, Anthropic's Claude, and Oobabooga (local) — it offers the flexibility to run different models for different agents based on your specific needs.

Whether you're a newbie looking for a user-friendly entry point or a seasoned developer aiming to build complex cognitive architectures, this framework has you covered.

Our database-agnostic framework is designed for seamless extensibility. While ChromaDB is our go-to database, integration with other databases is straight-forward, making it an ideal playground and solid foundation for various AI projects.

In summary, AgentForge is your beta-testing ground and future-proof hub for crafting intelligent, model-agnostic, and database-flexible autonomous agents.


Table of Contents

  1. Features
  2. Documentation
  3. Contributing
  4. Contact Us
  5. License

Features

  • Build Custom Agents And Cognitive Architectures Easily
  • Custom Tools/Actions
  • Customizable Agent Memory Management
  • Default Agents Ready For Use
  • LLM Agnostic Agents (Each Agent can call different LLMs if needed)
  • On-The-Fly Prompt Editing
  • OpenAI & Anthropic API Support
  • Open-Source Model Support (Oobabooga)
  • Rapidly Build & Test Cognitive Architectures (Multi-Agent Scripts)

Coming Soon

  • API Implementation
  • Knowledge Graphs

Documentation

Welcome to the AgentForge framework documentation. This comprehensive guide will support you whether you're just getting started or diving deep into custom configurations and advanced features. Here, you'll find detailed insights into the various components that make up our system.

Installation and Usage:

  • Getting Started with AgentForge: Begin your journey with a straightforward setup guide, covering everything from initial installation to launching your first bot.

Core Concepts:

  • Agents: Dive deep into the agents' world. Learn how they operate, respond, and can be customized.

  • LLM API Integration: Understand how AgentForge connects with various Large Language Model (LLM) APIs.

  • Settings: Delve into directives, memories, models, paths, and storage configurations – the fine-tuned engine settings that keep the system humming.

  • Personas: Add flair to the system's interactions. Make it more human-like and relatable.

  • Tools & Actions: The system's utility belt. Understand the tools available and how they can be choreographed into actionable sequences.

  • Utilities: Explore the array of utility functions and tools that supercharge the system's capabilities. (Note: Documentation not implemented yet)

Note: Our documentation is a living entity, continuously evolving. Some links or features may still be under development. We appreciate your patience and welcome your feedback to improve the documentation.


Contributing

Feel free to open issues or submit pull requests with improvements or bug fixes. Your contributions are welcome!

Special Note

We're on the lookout for a UI/UX collaborator who's passionate about open-source and wants to join the team to help develop a front-end for this framework. This isn't a job offer, but rather an invitation to be a part of something cool. Interested? We'd love to chat! (See the Contact Us section below for details.)


Contact Us

If you're keen on contributing or just want to reach out to us, here's how to get in touch:


License

This project is licensed under the GNU General Public License. See LICENSE for more details.

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

agentforge-0.1.32.tar.gz (687.3 kB view hashes)

Uploaded Source

Built Distribution

agentforge-0.1.32-py3-none-any.whl (701.8 kB view hashes)

Uploaded Python 3

Supported by

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