The MCP Ecosystem Platform aggregates 11 specialized MCP servers into a unified development and automation platform. It provides AI-powered code analysis, security scanning, browser automation, and intelligent workflow orchestration, all accessible through a single startup script. To begin, clone the repository and run the one-command startup script, which installs dependencies, validates prerequisites, and boots all services in the proper order. Once running, you can interact with the frontend for orchestration and use the backend API to manage MCP pipelines, monitor health, and inspect status across the entire MCP stack. The platform is designed to chain MCP servers to execute complex tasks, enabling real-time code intelligence, security monitoring, and automated browser-driven research as part of your development workflow.

The available tools cover: AI-driven code processing via groq-llm and openrouter-llm, browser automation with browser-automation and real-browser, security components like api-key-sniffer and network-analysis, as well as enhanced filesystem and git capabilities through enhanced-filesystem and enhanced-git. The Simple-warp/terminal server provides automation and commands within a unified dashboard, while kiro-tools ties together core operations across the stack. You can compose workflows by linking these servers so one task flows through analysis, security checks, data gathering, and automated actions.

Prerequisites:

• Python 3.11+ (and virtual environments support)
• Git
• Optional: Node.js for frontend development (not required to run the MCP server core)

Installation steps:

1. Clone the repository: git clone https://github.com/turtir-ai/mcp-ecosystem-platform.git cd mcp-ecosystem-platform

2. Optional: Create and activate a Python virtual environment: python -m venv venv

   Windows

   venv\Scripts\activate

   Linux/Mac

   source venv/bin/activate

3. Install backend dependencies (if using the provided one-command startup):

   From project root

   python -m pip install -r requirements.txt

4. Prepare environment (optional but recommended):

   • Copy environment template if present and customize API keys and secrets: cp .env.example .env
   • Edit .env with your credentials and configuration

5. Run the one-command startup script to boot all MCP services: python start-dev.py

6. Access the platform:

   • Frontend: http://localhost:3000
   • Backend API: http://localhost:8001
   • MCP Manager: http://localhost:8009

Notes:

• If you prefer manual startup, you can start the backend with uvicorn, the frontend separately, and the MCP Manager via its own entrypoint, but the one-command script is recommended for correctness and dependency handling.

Tips and common issues:

• Ensure your environment variables (.env) or the ENV block are properly set for API keys and external services.
• The platform runs multiple services; ensure your system has sufficient resources (CPU, memory) to avoid startup failures.
• If the UI cannot connect to the backend, verify that the backend API is listening on port 8001 and that there are no firewall or proxy blocks.
• When upgrading dependencies, re-run the startup script to re-install and reconfigure as needed.
• The 11 MCP servers are designed to be composed into workflows; use the Unified Control Interface to visualize and orchestrate complex tasks across servers.
• For debugging, consult the health endpoints (e.g., /health) and MCP status endpoint to confirm all components are healthy.