20250223 MCP Client

Provides multi-provider AI context management with tool coordination and scalable APIs for developers.

python

GitHub Stars

python

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "mark850409-20250223_mcp-client": {
      "command": "python",
      "args": [
        "-m",
        "mcpcli.web",
        "--openai-key",
        "sk-xxx...",
        "--anthropic-key",
        "sk-ant-xxx...",
        "--google-key",
        "xxx..."
      ],
      "env": {
        "OLLAMA_HOST": "http://localhost:11434",
        "GOOGLE_API_KEY": "YOUR_GOOGLE_API_KEY",
        "OPENAI_API_KEY": "YOUR_OPENAI_API_KEY",
        "ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY"
      }
    }
  }
}

Model Context Provider (MCP) is a versatile AI development tool that unifies context management, tool coordination, and multi-model orchestration. It helps you build more capable AI assistants by efficiently handling long contexts, invoking a rich set of tools, and balancing between multiple AI providers to deliver reliable and scalable interactions.

How to use

You run MCP through lightweight local interfaces or a web UI, connect to the MCP servers you want to use, and start chatting with intelligent assistants that can reason over a broad context and call tools as needed. You can pick different AI providers, configure models, and use built-in tools for code analysis, document handling, data retrieval, and external API calls. Start a session, share your prompts, and MCP takes care of context history, prompts, and tool orchestration to keep conversations coherent and productive.

How to install

Prerequisites: you need Python installed on your system. You may also use the UV runtime to manage MCP servers.

pip install uv
uv pip install fastapi uvicorn gradio google-generativeai ollama

uv sync

Additional sections

You can run MCP via multiple interfaces. The following local commands start the MCP web UI, the Gradio graphical interface, and a command-line interface respectively. Each method accepts your API keys and uses the same underlying MCP server logic.

# Start the FastAPI web interface
python -m mcpcli.web --openai-key sk-xxx... --anthropic-key sk-ant-xxx... --google-key xxx...

# Start the Gradio UI
python -m mcpcli.web.gradio_app --openai-key sk-xxx... --anthropic-key sk-ant-xxx... --google-key xxx...  --port 8082

# Start the CLI interface
python -m mcpcli.cli --server github --provider ollama --model llama3.2 --openai-key sk-xxx... --anthropic-key sk-ant-xxx... --google-key xxx...

Security and usage notes

Set and manage your API keys securely. Avoid exposing keys in unsafe environments. Monitor usage to stay within quotas and enable proper logging for troubleshooting.

Troubleshooting tips

If you encounter connectivity or key issues, verify that your keys are correctly set, check that the target MCP server is reachable, and review logs for detailed errors.

Built by

VeilStrat

AI signals for GTM teams