Skip to content
Vendia

Vendia MCP Gateway FAQ

Quick Start Questions

What is Vendia MCP Gateway?

Vendia MCP Gateway connects AI applications like Claude (an AI assistant by Anthropic) to your data through the Model Context Protocol (MCP). It’s a fully managed service available in three tiers:

  • Free Tier: Connect AI applications to your existing Amazon S3 buckets, remote systems via API Integrations, or remote systems via MCP Server Integrations. Once the connections are made access controls provide the ability to control what assets within those systems are available to AI.
  • Pro Tier: Everything in the Free Tier with increased limits, support for teams and deployment in region of choice.
  • Enterprise Tier: Everything in Pro plus access to operations data, analytical tables, file management, and Vendia workflows. Enterprise also supports replication of data to additional regions for data sovereignty.

For detailed information, see our Overview.

Do I need to install or deploy anything?

No installation required! Vendia MCP Gateway runs as a managed service. The free tier offering is deployed in us-east-1. With pro and enterprise tiers the customer can choose to have theirs gateways deployed in their region of choice.

Which AI applications is this compatible with?

Any application supporting the Model Context Protocol, including Claude, Mistral AI: Le Chat, OpenAI-compatible apps, and custom AI applications. See the Overview for compatibility details.

Since this is an open standard, it will work with any future AI applications that implement the MCP protocol as well!

Where do I find my MCP Gateway URL?

The location depends on which Vendia service tier you’re using:

Free & Pro Tier Users: Your MCP Gateway URL is provided on the getting started page. Copy this URL - you’ll need it for your AI application configuration.

Enterprise Tier Users: Go to your Vendia project dashboard → Project Settings → Resources → look for “MCP Gateway URL”.

This URL works for all connection methods (remote, local proxy, or programmatic). See our Getting Started Guide.

Do I need to install anything on my client system?

For modern AI clients like Claude that support remote MCP connections, no installation is required. Simply add the MCP Server URL directly in your client.

For older clients that don’t support remote connections, you’ll need Node.js (v22 LTS or later) and npm to use the local proxy method. See Advanced Connection Methods - Local Proxy.

Can I use this with multiple AI applications at once?

Yes! Each AI application authenticates independently and can access data according to its permissions.

What are some use cases?

Intelligent Business Operations

  • Supply Chain Intelligence: AI agents that can query real-time inventory, orders, and logistics data across your partner network
  • Financial Analysis: LLMs accessing distributed financial data for automated reporting and insights
  • Compliance Monitoring: AI-powered compliance checks across multi-party regulated data

Customer Experience Enhancement

  • Contextual Support: AI assistants with access to real-time customer data, order history, and partner interactions
  • Personalized Recommendations: ML models that can access comprehensive user profiles across partner ecosystems
  • Dynamic Pricing: AI systems that can access real-time market data, inventory levels, and demand signals

Data Product Monetization

  • AI-Ready Data Products: Transform your Vendia data shares into AI-accessible services
  • Partner AI Enablement: Allow partners to integrate their AI tools with your shared data products
  • Marketplace Integration: Connect your data to AI marketplaces and platforms

Security Questions

What authentication methods are available?

Vendia MCP Gateway supports three authentication methods to accommodate different types of AI applications and integration tools:

  1. OAuth Flow (Recommended): For interactive AI applications (e.g., Claude, ChatGPT, and Mistral AI: Le Chat)
  2. Service Credentials via Headers: Simplified authentication for no-code tools (e.g., OpenAI Agent Builder, Zapier, Make) that cannot perform OAuth flows
  3. Programmatic Token Exchange: For custom applications requiring full control over authentication

See the Authentication Methods guide for detailed information about each method, security considerations, and when to use each approach.

What permissions does the MCP client have to access my data?

The MCP client inherits permissions based on your authentication method:

  • OAuth (login credentials): The client has access through your default role.
  • API credentials (service credentials or programmatic): The client has permissions of the specific role linked to those API credentials.

This follows Vendia’s Role-Based Access Control (RBAC) model. To limit MCP access, you can create API credentials with a restricted role.

Should I use service credentials or OAuth for my integration?

Use OAuth (recommended) when:

  • Your AI application supports interactive authentication flows
  • You want the highest level of security
  • Users should authenticate with their own credentials

Use service credentials when:

  • Your no-code tool cannot perform OAuth flows (e.g., OpenAI Agent Builder, Zapier, Make)
  • Simplified authentication is required for low-code platforms
  • Your integration tool only supports custom headers

Use programmatic token exchange when:

  • Building custom AI applications
  • You need full control over token management
  • You want to implement custom authentication logic

For detailed guidance and security considerations, see the Authentication Methods guide.

How does Vendia MCP Gateway ensure the security of my data and AI interactions?

We offer enterprise-grade security with OAuth2, encryption, and fine-grained access control.

Fully supports the MCP authorization specification.

Details in Overview - Authentication & Security.

What security features are included?

  • Secure Architecture: Leverages Vendia’s foundational security model, including end-to-end encryption
  • Fine-Grained Access Control: Granular permissions aligned with your existing Vendia workspace access policies and RBAC configuration.
  • Audit Trail: Complete visibility into all data access patterns and AI interactions through Receipts

Performance Questions

Are there any usage limits for using Vendia’s MCP Gateway?

The MCP Gateway handles typical AI workloads efficiently. For specific limits or high-volume use cases, contact Vendia Support.

Can I use this for bulk data processing?

The MCP Gateway is optimized for AI use cases rather than bulk processing. For large-scale operations, consider using Vendia’s native APIs directly.

Technical Questions

How is Vendia MCP Gateway architected to deliver its robust capabilities?

Vendia MCP Gateway runs as a fully managed service with enterprise-grade security and OAuth2 authentication. See Overview - Authentication & Security for security details.

Getting Help

Can you provide more detailed setup instructions

See our comprehensive Getting Started Guide with:

  • Remote MCP setup (recommended for Claude)
  • Local proxy setup for older clients
  • Programmatic access for AI applications

Can you provide more technical details about the MCP Gateway?

Check out the Overview for feature explanations, use cases, and available tools.

Who should I reach out to if more technical support is needed?

Contact Vendia Support for assistance with your specific setup or use case.

Where can I learn about Vendia concepts?

Still have questions? The Vendia team is here to help! Contact Vendia Support for assistance with your specific use case.