Build + Deploy ServiceNow Scripts with AI in 5 Minutes! | FastMCP + Gradio Full Tutorial & Live Demo

Video Description

Need to crank out Business Rules or Client Scripts *instantly*—and ship them straight to your ServiceNow dev instance? In this step-by-step video I’ll show you how to: 1. Spin up a FastMCP server- that exposes ServiceNow CRUD tools over SSE/HTTP. 2. Install & launch- the Gradio-powered “Script Manager” UI that taps GPT-4o for code generation and inline AI edits. 3. Connect everything with one JSON stanza so Claude, Cursor, Windsurf—or your own LLM client—can call those same tools. 4. Walk through a **live end-to-end demo**: generate, edit, deploy, and update scripts… all in under five minutes! --- Commands demo’d in the video # clone repo & install deps git clone https://github.com/yvain13/MCP-Use-with-SN cd MCP-Use-with-SN python -m venv venv && source venv/bin/activate pip install -r requirements.txt # create .env with your keys cp .env.example .env # (edit and add OPENAI_API_KEY + SN creds) # terminal 1 – start FastMCP server python servicenow_br_mcp_server.py # terminal 2 – launch Gradio UI python script_manager.py ``` Optional: hook up external LLM clients ```json { "mcpServers": { "remote-example": { "command": "npx", "args": ["mcp-remote", "http://localhost:9123/sse"] } } } ``` --- What you’ll learn - Why FastMCP makes ServiceNow APIs “LLM-friendly” - Crafting one-shot prompts so GPT returns perfect JSON every time - Sharing a persistent asyncio loop to avoid ‘event loop closed’ errors - Securing secrets with a local `.env` (and scrubbing them from Git history!) --- 💬 Questions? Drop them in the comments—I answer every one. 👍 Like & subscribe if this saved you hours of manual scripting! --- **#ServiceNow #AI #GPT4o #FastMCP #Gradio #Automation #DevOps****🎥 Title**