Hierarchical multi-agent system for complex task solving
Top 33.3% on sourcepulse
DeepResearchAgent is a hierarchical multi-agent system designed for complex research and general task solving, coordinating specialized agents under a top-level planner. It offers automated task decomposition, information analysis, web research, and tool integration for users needing sophisticated AI-driven workflows.
How It Works
The system employs a two-layer architecture: a top-level planning agent decomposes tasks and assigns them to specialized lower-level agents (Deep Analyzer, Deep Researcher, Browser Use, General Tool Calling). This hierarchical approach enables dynamic coordination and efficient execution of complex, multi-step processes, leveraging asynchronous operations for improved performance.
Quick Start & Requirements
poetry install
or make install
(requires Python 3.11+).playwright install chromium --with-deps --no-shell
may be needed for browser tools..env
file with API keys (e.g., Google Gemini).Highlighted Details
Maintenance & Community
Licensing & Compatibility
Limitations & Caveats
The project acknowledges shortcomings in dynamic interactions on Google Street View and video-audio analysis, with ongoing optimization efforts. The README does not specify a license, which may impact commercial adoption.
1 week ago
Inactive