A lightweight and developer-friendly agent framework for building intelligent, autonomous agents.

• Lightweight Architecture: Minimal overhead and fast execution
• Developer-Friendly API: Simple, intuitive interface for creating agents
• Modular Design: Extensible components for custom functionality
• Multi-Agent Support: Coordinate multiple agents working together
• Flexible Configuration: Easy setup with various configuration options
• Event-Driven: Responsive to external stimuli and internal state changes
• Extensible: Plugin system for adding new capabilities

pip install mini-agent

from mini_agent import Agent, Task

# Create a new agent
agent = Agent(name="my-agent", config={"verbose": True})

# Define a simple task
task = Task(description="Process data and return results")

# Execute the task
result = agent.execute(task)
print(result)

Agents are the primary building blocks that can perceive their environment, make decisions, and take actions.

Tasks represent units of work that agents can execute, with defined inputs and expected outputs.

Agents can maintain short-term and long-term memory to store information between interactions.

Agents can utilize various tools for external interactions and computations.

The framework supports configuration through:

• Environment variables
• Configuration files (JSON/YAML)
• Programmatic setup

• Automated data processing
• Intelligent assistants
• Workflow automation
• Multi-agent coordination
• Decision support systems

We welcome contributions from the community! Please see our Contributing Guide for more details.

This project is licensed under the MIT License - see the LICENSE file for details.

For support, please open an issue in the GitHub repository or contact the maintainers.