Multi-agent-system-with-MCP-tool-integration

PKMadhavan/Multi-agent-system-with-MCP-tool-integration

★ 0 stars Python AI/LLM Updated today

Multi-agent system with MCP tool servers and A2A orchestration — an orchestrator routes tasks to specialist agents over A2A, each backed by sandboxed MCP tools (RAG, calculator, web search), with guardrails and a reproducible evaluation harness.

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "multi-agent-system-w": {
      "command": "uvx",
      "args": [
        "multi-agent-system-with-mcp-tool-integration"
      ]
    }
  }
}

Or install with pip: pip install multi-agent-system-with-mcp-tool-integration

README Excerpt

A production-shaped multi-agent system: an **orchestrator agent** routes tasks to two **specialist agents** over the **A2A (agent-to-agent) protocol**; each specialist calls out to sandboxed tools exposed as **MCP (Model Context Protocol) servers**. The domain is cleantech & climate-policy Q&A (carbon pricing, renewable incentives, grid storage,

Multi-agent-system-with-MCP-tool-integration

Quick Install

README Excerpt

Related AI/LLM Servers