AI series · Part 4: Agentic Automation
MCP (Model Context Protocol): The USB-C Cable for AI
MCP is an open standard for connecting AI apps to tools and data. Build a connector once and every MCP-compatible app can use it. Here's what an MCP server offers.
MCP (Model Context Protocol) is an open standard from Anthropic for connecting AI apps to tools and data. You build a connector once, and every MCP-compatible app can use it.
It's the USB-C cable for AI. Before USB-C every phone needed its own charger; now one cable fits all.
What an MCP server offers
| Offers | What it is | Example |
|---|---|---|
| Tools | Actions the AI can take | apply_leave |
| Resources | Data the AI can read | Policy documents |
| Prompts | Ready-made instructions | "Summarise this policy for a new joiner" |
The ecosystem
There are MCP servers for GitHub, Slack, Google Drive, Jira, Postgres and many more. Install the GitHub one once, and Claude Desktop, Cursor and VS Code can all use it.
MCP in our HR assistant
Our reference build exposes its HR tools (policy search, employee lookup, apply leave) as an MCP server. Any MCP app can use the HR assistant without a single line of extra integration code.
MCP standardizes tool and data connections, so instead of building a custom integration for every app and tool combination, each tool is built once as an MCP server.
- The short version
Next in the GenAI series: Letting AI Agents Use the Command Line, Safely
Start from the top: GenAI Quick Answers and Series Guide
We deploy these patterns across production engagements in fintech, healthcare, and eCommerce. Talk to a senior engineer about how they apply to your platform.