|
# Autonumous Agents: Multi-LLM |
|
|
|
<p align="center"> |
|
<a href=""><img src="docs/resources/logo-creatus.png" alt="Creatus Autonumous Agents logo: A Framework for Automatic Agent Generation." width="300px"></a> |
|
</p> |
|
|
|
<p align="center"> |
|
<b>Generate different roles for LLMs to form a collaborative entity for complex tasks.</b> |
|
</p> |
|
|
|
<p align="center"> |
|
An Automatic Agents Generation Framework based on LLMs. This program, driven by LLMs, autonomously generates multi-agents to achieve whatever goal you set. |
|
</p> |
|
|
|
<p align="center"> |
|
<img width="600" src="docs/resources/framework2.jpg"> |
|
</p> |
|
|
|
## <a name="updates"/> π Updates |
|
- **2023.08.30**: Adding a custom agent collection, AgentBank, allows you to add custom agents. |
|
<p align="center"> |
|
<img width="600" alt="The execution process of AutoAgents." src="docs/resources/process.jpg"> |
|
</p> |
|
|
|
## π Features |
|
- **Planner**: Determines the expert roles to be added and the specific execution plan according to the problem. |
|
- **Tools**: The set of tools that can be used, currently only compatible with the search tools. |
|
- **Observers**: Responsible for reflecting on whether the planner and the results in the execution process are reasonable, currently including reflection checks on Agents, Plan, and Action. |
|
- **Agents**: Expert role agents generated by the planner, including name, expertise, tools used, and LLM enhancement. |
|
- **Plan**: The execution plan is composed of the generated expert roles, each step of the execution plan has at least one expert role agent. |
|
- **Actions**: The specific actions of the expert roles in the execution plan, such as calling tools or outputting results. |
|
|
|
## Acknowledgements |
|
The system, action bank and role bank of this code base is built using [MetaGPT](https://github.com/geekan/MetaGPT). The basic framework is based on [AutoAgents](https://github.com/Link-AGI/AutoAgents) |
|
|
|
|