English
mahilo is a powerful AI proxy integration platform designed to connect AI proxy from different frameworks together for real-time communication and human supervision. It supports a variety of popular proxy frameworks such as LangGraph, Pydantic AI, etc. by providing framework-independent communication protocols, while allowing connection to proprietary proxy through APIs. The platform emphasizes intelligent collaboration, organization-level strategy management and human-centric design to ensure that human control is maintained while automation. The emergence of mahilo provides flexible solutions for building complex multi-agent systems, suitable for a variety of application scenarios, from content creation to emergency response. Currently, mahilo has 251 stars on GitHub and has more than 500 PyPI downloads per month, showing its popularity in the developer community. mahilo is mainly aimed at developers and enterprise users, helping them quickly build and deploy multi-agent systems to improve work efficiency and innovation capabilities.
Demand population:
" mahilo is mainly aimed at developers and enterprise users, especially those who need to build and manage multi-agent systems. It is suitable for those who want to integrate AI agents from different frameworks into a unified platform, enabling efficient collaboration and real-time communication. In addition, for scenarios where human supervision and control are needed in the process of AI automation, mahilo also provides the ideal solution. Whether it is content creation, emergency response or commercial applications, mahilo can help users quickly build and deploy complex multi-agent systems, improving work efficiency and innovation capabilities."
Example of usage scenarios:
Story Weaver: A multi-player story creation game, where users collaborate with AI agents to create stories, AI intelligently integrates narratives, supports real-time collaborative creation, shared context management and multi-player AI interaction.
911 Emergency Response: In the emergency response scenario, AI agents can coordinate resources from all parties, respond quickly and provide support, and improve emergency response efficiency.
RentMate AI: A real estate matching application. AI agents help users find suitable housing sources and improve rental efficiency through intelligent matching and information sharing.
Product Features:
General proxy integration: supports AI proxy connecting to multiple frameworks such as LangGraph, Pydantic AI, and can connect to proprietary proxy through API.
Real-time communication: Provides instant voice and text chat capabilities for integrated agents, supporting real-time communication and seamless human-machine interaction between agents.
Intelligent collaboration: AI agents can independently share context and information, realizing cross-frame information exchange and automatic proxy query.
Organization-level policy management: Centralized management of policies to ensure consistent behavior and security controls for all integration agents.
Anthropocentric design: maintain human control while AI processes complex interactions, contact humans only when necessary, and allow humans to interfere in AI decision-making.
Multi-agent architecture: supports the construction of complex proxy systems with flexible communication modes, including hierarchical and point-to-point modes.
Multi-user interaction: supports multiple users to interact with AI agents at the same time, with complex shared context management and configurable inter-agent communication modes.
Flexible development interface: Provides tools such as BaseAgent class and AgentManager to facilitate developers to define agents, create agent managers and start WebSocket servers.
Tutorials for use:
1. Define the proxy: Use the BaseAgent class or create a proxy through integration, such as using LangGraphAgent to connect to the proxy for the LangGraph framework.
2. Create a proxy manager: Add the defined proxy to the AgentManager and treat it as a team.
3. Start the WebSocket Server: Create and run the AgentWebSocketServer to enable real-time communication between agents.
4. Connect to the client: Use client scripts to connect to the WebSocket server, specify the agent name to start interaction, for example, connect to the buyer agent through the command `$ python client.py --agent-name buyer_agent`.
