Step 6: Connect Any AI

Goal: Connect any MCP-native AI to your governed data. This is the "Bring Any AI" proof.

---

What is MCP?

The Model Context Protocol (MCP) is the industry standard for connecting AI to data sources.

Think of it like HTTP for AI:

• HTTP lets browsers talk to web servers
• MCP lets AI systems talk to data sources

OpenIndustrial exposes your governed queries via MCP. Any MCP-native AI can connect - today.

---

Get Your MCP Credentials

1. In your workspace, click APIs Menu
2. Select MCP Configuration
3. Copy your MCP server URL:

   https://mcp.openindustrial.co/v1
   

4. Generate a Bearer Token (same process as JWT generation in Step 5)

---

Connect Claude Desktop

Claude Desktop natively supports MCP. Here's how to connect:

1. Find the Config File

OS	Location
macOS	~/Library/Application Support/Claude/claude_desktop_config.json
Windows	%APPDATA%\Claude\claude_desktop_config.json

2. Add OpenIndustrial

{
  "mcpServers": {
    "open-industrial": {
      "url": "https://mcp.openindustrial.co/v1",
      "headers": {
        "Authorization": "Bearer YOUR_JWT_TOKEN"
      }
    }
  }
}

Replace YOUR_JWT_TOKEN with your bearer token from the previous step.

3. Restart Claude Desktop

Close and reopen Claude Desktop. Look for the MCP connection indicator.

4. Test It

Ask Claude:

"What's the average temperature from my brewery data?"

Claude now queries your governed data directly.

---

Connect VS Code Copilot

VS Code Copilot can connect via MCP extensions.

1. Install MCP Extension

Search for "MCP" in VS Code extensions and install a compatible client.

2. Configure the Connection

Add to your workspace .vscode/settings.json:

{
  "mcp.servers": {
    "open-industrial": {
      "url": "https://mcp.openindustrial.co/v1",
      "headers": {
        "Authorization": "Bearer YOUR_JWT_TOKEN"
      }
    }
  }
}

3. Query Your Data in Code

Now Copilot understands your industrial data context:

// Copilot can autocomplete based on your governed data schema
const breweryData = await openIndustrial.query('temperature-avg');

---

Connect AI App Makers

Build apps that query governed data with Replit Agent or similar tools.

Replit Agent Example

1. Start a new Replit project

2. In the AI chat, explain:

   "I have an MCP server at https://mcp.openindustrial.co/v1 with brewery data. Build me a dashboard showing temperature trends."

3. Replit Agent will scaffold an app that queries your governed data

---

Available MCP Tools

When AI systems connect via MCP, they access these tools:

Tool	Description
warm-queries.catalog	List all available governed queries
warm-queries.run	Execute a specific query by name

AI systems can discover your queries and run them - all within the governance framework.

---

What Just Happened?

The AI you connected didn't need special integration. It just speaks MCP. That's the standard we built on.

---

What You Accomplished

Category Metric	Your Result
Time to First Governed Query	~5 minutes
Time to First AI Connection	~10 minutes
Governance established	Every query logged
AI portability	Any MCP-native AI works

---

The Category Difference

Traditional Platforms	OpenIndustrial Control Plane
Lock you to their AI	Connect ANY MCP-native AI
Proprietary protocols	Industry standard (MCP)
Switch costs compound	Fire us, keep your connections
Their moat is lock-in	Your moat is YOUR investment

---

What's Next?