1. Pipe17 Help Center
  2. Getting Started
  3. Initial Configuration & Features

Pipe17 MCP Server

John Shao

What is an MCP Server?

MCP (Model Context Protocol) is an open protocol that enables secure, standardized connections between AI assistants and external data sources or tools. An MCP server acts as a bridge between your AI assistant (like Claude) and your business systems, allowing the AI to access real-time data and perform actions on your behalf.

Think of it as giving your AI assistant secure, controlled access to your business tools - similar to how you might give a human assistant access to your company systems, but with precise control over what they can see and do.

 

Why Use the Pipe17 MCP Server?

Key Benefits

Accelerate Operations

  • Instantly query order statuses, inventory levels, and shipment tracking
  • Bulk update products, process orders, and manage inventory with natural language commands
  • Generate complex reports in seconds instead of minutes

AI-Powered Automation

  • Let AI handle repetitive tasks like order processing and status updates
  • Create shipping labels and manage fulfillments through conversational commands
  • Automate data entry and reduce manual errors

Intelligent Insights

  • Ask complex questions about your business in plain English
  • Get instant analysis of sales trends, inventory patterns, and operational metrics
  • Receive AI-powered recommendations for inventory optimization and order management

Seamless Integration

  • Works with any MCP-compatible AI assistant (Claude, and others coming soon)
  • No coding required - just configure and start using
  • Maintains your existing Pipe17 permissions and security settings

Natural Language Interface

  • No need to remember complex API endpoints or query syntax
  • Simply ask: "Show me all pending orders from last week" or "Update inventory for SKU-123"
  • The AI understands context and can handle multi-step operations

Example Use Cases

  • Order Management - "What's the status of order #12345?" or "Show me all orders with status 'readyForFulfillment'"
  • Inventory Tracking - "Show me inventory levels for SKU-123" or "List products with less than 10 units in stock"
  • Fulfillment Status - "Has order WOO-456 been fulfilled?" or "Show all fulfillments from last week"
  • Customer Queries - "Find customer details for email@example.com" or "List all customers created this month"
  • Integration Management - "Show me all active integrations" or "Check the status of my Shopify connection"
  • Exception Handling - "What exceptions need attention?" or "Show me all unresolved inventory exceptions"
  • Reporting - "Calculate total revenue from fulfilled orders this month" or "Show order trends by status"

 

How It Works

The Pipe17 MCP Server provides your AI assistant with:

  • Secure Access - Uses your existing Pipe17 authentication
  • Real-time Data - Connects directly to live Pipe17 systems
  • Controlled Permissions - Respects your account's access levels
  • Audit Trail - All actions are logged and traceable

 

Setting Up AI Assistants

Pipe17 supports multiple MCP-compatible AI assistants. Setup depends on which assistant you're using and whether it supports OAuth2 or API Key authentication.

  • OAuth2 is the recommended method. It provides secure, token-based access with automatic token refresh. Access scopes align with your Pipe17 account permissions.
  • API Key authentication is supported for CLI-based tools like Gemini. Use with caution and rotate keys regularly.

💡 Before you begin: You’ll need your Pipe17 Organization Key. To find it:

  1. Log into your Pipe17 account
  2. Go to Settings > Profile
  3. Copy your Organization Key from the top of the page

 

ChatGPT (OAuth2 Only)

If you have a ChatGPT Pro or Plus plan, you can add Pipe17 as a connector using developer mode.

Setup Steps

  1. Go to chatgpt.com
  2. Click your profile icon (bottom left) > Settings > Connectors
  3. Scroll to the bottom > Expand Advanced Settings
  4. Enable Developer Mode
  5. Go back to Browse Connectors > Click Create
  6. Configure with:
    • Name: Use Pipe17 – [YourOrgName]
      (You must reference this name in prompts to trigger the connector)

    • MCP Server URL:

      https://mcp.pipe17.com/mcp/YOUR_ORG_KEY

      Replace YOUR_ORG_KEY with your actual Pipe17 organization key (see below)

    • Authentication: Keep set to OAuth2
    • Select "I trust this application"
  7. Click Create to finish

 

Claude (OAuth2 Only)

Claude connects to Pipe17 using OAuth2, and setup requires action from both admins and end users.

Step 1: Admin Setup

Before users can connect Claude to Pipe17, an admin must create a custom connector in Claude:

  1. Go to claude.ai
  2. Click your profile icon (bottom left) > Settings
  3. Navigate to Connectors
  4. Click Add custom connector
  5. Configure with the following:
    • Name: Use your organization name (e.g., Pipe17 – Acme)
      (Claude will use this name to match prompts to connectors)

    • MCP Server URL:

      https://mcp.pipe17.com/mcp/YOUR_ORG_KEY

      Replace YOUR_ORG_KEY with your actual Pipe17 organization key (see below)

    • Authentication: Leave set to OAuth2
  6. Click Connect to start the OAuth flow and save the connector

Step 2: User Authorization

Once the connector is set up:

  • Individual users can authorize it directly within Claude
  • Each user's access level will match their Pipe17 permissions

 

Gemini CLI (API Key or OAuth2)

Gemini supports either API Key or OAuth2 authentication. Configuration is done in your local settings file.

Option 1: Using an API Key

  1. Generate an API key in your Pipe17 account:
    • Go to Settings > API Keys
    • Create a new key or copy an existing one
  2. Locate your Pipe17 org key (see below)
  3. Open your local file: ~/.gemini/settings.json
  4. Add the following block:
{
  "mcpServers": {
    "pipe17": {
      "httpUrl": "https://mcp.pipe17.com/mcp/YOUR_ORG_KEY",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Replace:

  • YOUR_ORG_KEY with your Pipe17 organization key
  • YOUR_API_KEY with your generated API key

Option 2: Using OAuth2 (If Supported)

If your Gemini version supports OAuth2:

  • Follow Gemini’s OAuth2 configuration steps
  • Use the same Pipe17 MCP Server URL
  • OAuth2 will prompt for account authorization and respect permission scopes

 

Security Best Practices

  • Keep your API keys secure. Do not share or expose them in code
  • Use OAuth2 whenever possible for secure token handling
  • Rotate API keys regularly to minimize exposure risk
  • Monitor account activity to detect unauthorized access

 

Available Features

Once connected, your AI assistant can access a wide range of operational and intelligence features via the MCP Server:

Core Operations

  • Orders – Query by status, date, external ID, and more
  • Products – Search and retrieve product details and variants
  • Inventory – View real-time inventory across locations
  • Fulfillments – Track fulfillment status and details
  • Shipments – Check shipment progress and carriers
  • Customers – Look up customer records and order history
  • Locations – Review and manage warehouse and fulfillment locations

Business Intelligence

  • Exceptions – View and manage system exceptions requiring attention
  • Events – Track system events and changes
  • Automations – Monitor automation rules and execution
  • Order Routing – See how orders are routed for fulfillment
  • Jobs – Review background job statuses and results

System Reference & Metadata

  • Knowledge Base – Access comprehensive Pipe17 documentation
  • Schema Information – Get data structure and field definitions
  • Entity Mappings – Understand data transformations and mappings
  • Integration Status – Monitor connected systems and their health

 

Troubleshooting & Support

Common Issues

  • Connection Failed – Verify your organization key is correct
  • Authentication Error – Check that your API key is valid or OAuth2 tokens are properly configured
  • Permission Denied – Ensure your Pipe17 account has the necessary permissions for the requested operations
  • Rate Limiting – If you receive a 429 error, reduce request frequency; the MCP server applies rate limits to protect performance

Getting Help

If issues persist:

  • Confirm that your organization key is entered correctly
  • Double-check your authentication method and credentials
  • Contact Pipe17 Support with the following:
    • Your organization key
    • A copy of the error message or logs

Additional Resources

Articles in this section

Was this article helpful?
1 out of 1 found this helpful

Comments

0 comments

Please sign in to leave a comment.

Have more questions?
Submit a request
Share it, if you like it.