This MCP server setup exposes lightweight capabilities that enable an AI agent to interact with the host environment. The Terminal Server lets the agent execute shell commands in an isolated workspace, providing command execution, output capture, and logging. The Memory Server offers persistent data storage so the agent can remember user preferences or prior interactions across sessions. Together with the client components, you can issue natural language requests that translate into terminal actions or memory operations, helping the agent manage local tasks and store context for later use. Use the client to request file operations, memory queries, or command execution, and rely on the server feedback to drive subsequent steps in your workflow.

Prerequisites:

• Python 3.9+
• Node.js 16+
• UV package manager (uv)
• Google API Key (optional for certain features)

Installation steps:

1. Clone the repository and navigate to it: git clone https://github.com/Techiral/mcp.git cd mcp

2. Set up Python environment and install dependencies: python -m venv venv

   Linux/macOS

   source venv/bin/activate

   Windows

   venv\Scripts\activate pip install -r requirements.txt

3. Configure environment variables (example):

   • Create and populate client config in clients/mcp-client/.env GOOGLE_API_KEY=your_api_key_here THEAILANGUAGE_CONFIG=clients/mcp-client/theailanguage_config.json

4. Install Node.js servers (for memory/filesystem as needed): npm install -g @modelcontextprotocol/server-memory @modelcontextprotocol/server-filesystem

5. Start the MCP client (which will coordinate the servers): python clients/mcp-client/langchain_mcp_client_wconfig.py

Notes:

• Ensure paths in the configuration are correct for your environment.
• Restart services after making configuration changes.

Tips and common issues:

• Absolute paths help avoid working-directory issues when launching servers.
• Ensure MEMORY_FILE_PATH points to a writable location with proper permissions.
• If the terminal or memory servers don’t start, check that uv and npx.cmd are available in your PATH and that the specified Python script paths are correct.
• When updating configurations, restart the MCP client to apply changes.
• Use the .env file to manage sensitive keys and configuration flags; do not commit these to source control.
• Verify that the required prerequisites (Python, Node, and UV) are installed and accessible from your shell.