-
Notifications
You must be signed in to change notification settings - Fork 676
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
feat: Multi-Agent System Based On The Role-Playing Module #346
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… into feature/multi-agent
… into feature/multi-agent
update enums and fix minor issues
for post-2023 developments
contextual parameters into subtask details
10 tasks
test_insight_agent.
instead of model
possibility and stability analysis
evaluate_role_compatibility.py and split_task.py
constructor
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Labels
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
We're excited to unveil our newly conceptualized framework design for a sophisticated multi-agent system (MAS), which promises to enhance task automation and process control through collaborative agent interaction.
Our framework's core is constructed around several primary features, ensuring a comprehensive and detailed approach to task management:
Task Split Feature: This functionality dissects complex tasks into manageable subtasks, utilizing a variety of parameters such as task prompts, roles with descriptions, and a subtasks-oriented graph. It lays the groundwork for parallel processing and execution pipelines, optimizing the system's overall performance.
Role Generation Feature: Key to the framework, this feature dynamically creates and assigns roles, encompassing the number, names, and descriptions, hence tailoring the system to handle specific tasks efficiently.
Process Control (SOP): Central to orchestrating the MAS, the standard operating procedure (SOP) guides agents through their assigned tasks. It integrates subtasks with dependencies, ensuring that each action by the role-playing agents aligns with the overarching task's objectives.
Action Agent: Acting as the executor within the framework, the Action Agent interfaces with APIs, tools, and commands to produce tangible execution results, informed by the selected roles and insights derived from data.
Task Assignment Feature: This assigns subtasks to the appropriate agents based on their roles and an evaluation of their performance, guaranteeing an optimal distribution of tasks.
Insight Generation Feature: By analyzing context and reference content, this feature provides agents with the insights necessary for informed decision-making and effective task execution.
Additionally, the framework includes specialized components like:
Subtask JSON: An organized data structure providing details of each subtask, including descriptions, dependencies, inputs, and outputs, enabling agents to understand and process their specific roles.
Env Agent: A crucial component that interacts with databases and utilizes tags to maintain and update the environment in which the agents operate.
Deductive Reasoner: This agent processes states to deduce logical outcomes, providing a result of reasoning that aids in decision-making processes.
Our design is not just a blueprint; it's a commitment to elevating the intelligence and efficiency of multi-agent systems. By fostering an environment where agents can learn, adapt, and cooperate, our framework is poised to transform task execution in complex operational landscapes.
As we move forward with the development, we are keen to collaborate with experts and enthusiasts in the field to refine and realize the full potential of this design. Stay tuned for updates as we progress in turning this framework into a functioning model that redefines collaborative task execution.
Types of changes
What types of changes does your code introduce? Put an
x
in all the boxes that apply:Implemented Tasks
Checklist
Go over all the following points, and put an
x
in all the boxes that apply.If you are unsure about any of these, don't hesitate to ask. We are here to help!