Skip to main content
This operational guide provides step-by-step instructions for running, configuring, and connecting the DSOM FastMCP Knowledge Server (tools/mcp/server.py) via Model Context Protocol (MCP) to AI IDEs such as Cursor, Claude Desktop, and VSCode.

⚑ 1. Fast Execution Protocol (Rule 16 uv Mandate)

The MCP server runs natively in Python using uv with PEP 723 inline script dependencies (mcp, fastmcp, pyyaml). Zero global installation required.

Launching the Stdio MCP Server:


πŸ”Œ 2. IDE Integration & Configuration Guides

A. Cursor IDE (.cursor/mcp.json or Settings -> MCP):

Add the following JSON configuration to Cursor under Settings > Features > MCP:

B. Claude Desktop (claude_desktop_config.json):

Add the server entry to %APPDATA%\Claude\claude_desktop_config.json (Windows) or ~/Library/Application Support/Claude/claude_desktop_config.json (macOS):

πŸ“Š 3. Exposed MCP Endpoint Matrix (Resources & Tools)

A. MCP Resources (6 Read-Only Protocol URIs)


B. MCP Tools (3 Executable Tools)


πŸ§ͺ 4. Troubleshooting & Verification

  1. Verify Python Environment: Ensure uv is installed and accessible on your PATH (uv --version).
  2. Test Server Startup:
  3. Verify Regression Suite:

πŸ”— 5. References


Deep State of Mind (DSOM) For My AI Protocol | Harisfazillah Jamel (LinuxMalaysia) | 2026-08-09 Standard: UK English | DBP-standard Bahasa Melayu Malaysia (Piawai) | GNU General Public License v3.0