Skip to content

setupguide.md

Latest commit

 

History

History
112 lines (90 loc) · 3.68 KB

setupguide.md

File metadata and controls

112 lines (90 loc) · 3.68 KB

MCP CLI System Configuration Guide

This guide outlines the steps to check your system configuration environment to ensure the MCP CLI operates correctly.

1. Python Environment

The MCP CLI requires Python 3.11 or newer.

How to check: Open your terminal and run:

python3 --version

Ensure the output shows Python 3.11.x or higher. If not, you may need to install or update your Python version.

2. Virtual Environment and Dependencies

It is highly recommended to use a virtual environment to manage project dependencies. This project uses uv for dependency management.

How to check/set up:

  1. Install uv (if not already installed): 
    curl -LsSf https://astral.sh/uv/install.sh | sh
    Or, if you have pipx: 
    pipx install uv
  2. Create and activate a virtual environment: Navigate to the project root directory (/Users/fanzhang/Documents/github/mcp-cli) and run: 
    uv venv
source .venv/bin/activate
  3. Install dependencies: 
    uv sync
    This will install all necessary packages listed in pyproject.toml and uv.lock.

3. Configuration Files

The MCP CLI relies on several configuration files.

  • server_config.json: Located in the project root, this file configures the MCP server connection. How to check: Verify its presence and content. A basic server_config.json might look like:

    {
  "mcp_server_url": "http://localhost:8000"
}

    Adjust mcp_server_url if your server is hosted elsewhere.

  • ~/.chuk_llm/providers.yaml: This file configures your LLM providers (e.g., OpenAI, Anthropic). How to check:

    cat ~/.chuk_llm/providers.yaml

    Ensure your desired providers are configured correctly with their respective API keys (if applicable). Example for OpenAI:

    openai:
  api_key: "sk-..."
  model: "gpt-4o"

  • ~/.chuk_llm/.env: This file can store environment variables, including sensitive API keys. How to check:

    cat ~/.chuk_llm/.env

    Ensure any API keys or other sensitive environment variables are correctly set here.

4. Environment Variables

Some configurations, especially API keys, might be set via environment variables.

How to check: You can check specific environment variables using echo:

echo $OPENAI_API_KEY
echo $ANTHROPIC_API_KEY
# etc.

Ensure these variables are set if your providers.yaml or other configurations rely on them.

5. Network Connectivity

The MCP CLI communicates with an MCP server and potentially external LLM provider APIs.

How to check:

  1. MCP Server Connectivity: If your mcp_server_url is http://localhost:8000, ensure your MCP server is running. You can try to ping it (though ping might not work for HTTP services, it's a quick check for host reachability):

    ping localhost

    A more direct check would be to use curl on the server's health endpoint if available, or simply try running an MCP CLI command that connects to the server.

  2. LLM Provider API Connectivity: Ensure you have internet access and that there are no firewall rules blocking access to LLM provider endpoints (e.g., api.openai.com, api.anthropic.com).

6. Running Diagnostics

The project includes diagnostic scripts that can help verify the setup.

How to run: Navigate to the diagnostics/ directory and run relevant scripts. For example:

cd diagnostics/
python3 provider_list_diagnostic.py
python3 mcp_server_diagnostic.py

These scripts can provide insights into your LLM provider and MCP server configurations.