A Multi-Agent platform that enables AI agents to collaborate effectively, helping you build powerful agent teams for solving complex tasks. It addresses the challenge of coordinating AI agents by providing frameworks for inter-agent communication, task planning, and autonomous decision-making, enabling users to build and deploy agent teams rapidly. The platform targets developers and researchers seeking to leverage collaborative AI for sophisticated problem-solving.

How It Works

The platform employs a modular, layered architecture for flexibility. It supports agent collaboration with role definition, task allocation, and multi-turn autonomous decision-making. AgentMesh integrates seamlessly with leading LLMs (OpenAI, Claude, DeepSeek) via a unified API and offers extensible tools like search, browser automation, and terminal execution.

Quick Start & Requirements

• Installation: Available via CLI, Docker, or Python SDK (pip install agentmesh-sdk).
  • CLI: Clone repo, pip install -r requirements.txt. Browser tools require pip install browser-use==0.1.40 and playwright install (Python 3.11+ recommended).
  • Docker: Use provided docker-compose.yml.
• Prerequisites: Python 3.7+ (3.11+ for browser tools), Linux/macOS/Windows, LLM API keys.
• Configuration: Edit config.yaml for model and agent settings.
• Execution: CLI: python main.py -t <team> -q "<question>". Docker: Interactive within the container. SDK: Programmatic use.
• Demo: https://github.com/user-attachments/assets/a0e565c4-94ef-4ddf-843d-a0c5aab640c6.

Highlighted Details

• Unified API for OpenAI, Claude, DeepSeek, Qwen LLMs.
• Built-in tools: calculator, current time, browser, Google Search, file save, terminal.
• Core features: Role definition, task allocation, multi-turn autonomous decision-making.
• Multi-platform support (CLI, Docker, SDK).

Maintenance & Community

• Contributions encouraged via PRs and issues.
• No explicit community channels, core maintainers, sponsorships, or roadmap details provided in the README.

Licensing & Compatibility

• No license information is explicitly stated in the provided README text.
• No compatibility notes for commercial use or closed-source linking are provided.

Limitations & Caveats

Features like remote heterogeneous agent communication, MCP protocol extensions, a WebUI, and local Ollama support are marked as "coming soon." Browser tools require Python 3.11+.