What is the Artificial Wit MCP server?

A single Model Context Protocol endpoint that exposes every API, knowledge base and tool you configure in Artificial Wit AI to any MCP-compatible client — Claude, Cursor, ChatGPT, Gemini and more.

Which AI clients can connect to the MCP endpoint?

Claude Desktop, Cursor, ChatGPT (via Actions / Custom GPTs), Gemini, and any open-source or custom client that supports the Model Context Protocol.

Is the MCP endpoint secure?

Yes. Access is gated by API keys and role-based permissions, credentials are stored securely, and every tool call is auditable.

Model Context Protocol

One MCP Server, every AI client

Connect Claude, ChatGPT, Cursor, Gemini and any MCP-compatible agent to your enterprise APIs, knowledge bases and workflows — through a single secure endpoint.

MCP Server Endpoint

Every configured API instantly becomes a Model Context Protocol tool — discoverable by Claude Desktop, Cursor, ChatGPT, and any MCP-compatible client through one secure endpoint.

Endpoint

https://api.artificialwit.com/mcp

Claude Desktop config

{
  "mcpServers": {
    "artificialwit": {
      "url": "https://api.artificialwit.com/mcp",
      "headers": { "x-api-key": "YOUR_API_KEY" }
    }
  }
}

Standardised tool interface across every LLM
Secure, permissioned execution per role
Plug & play with enterprise systems
Composable across multiple agents
Reusable credentials & global variables
Auditable, single endpoint to govern

Live Example

Connected in Claude as a Connector

Configured APIs appear instantly as tools with per-tool access control

Artificial Wit MCP Server connected in Claude showing custom connector and tool permissions for Assets, Vendors, Locations, Orders

Supported Integrations

Wire the Artificial Wit MCP server into the AI clients and automation platforms your team already uses.

Claude Desktop

Add Artificial Wit as a remote MCP server in Claude Desktop's config to expose every configured API and knowledge base as a native tool inside Claude.

Cursor

Wire the MCP endpoint into Cursor's settings so the IDE's AI can query your enterprise data, trigger workflows, and read internal docs without leaving the editor.

ChatGPT (Custom GPTs & Actions)

Use the MCP endpoint as an action surface for Custom GPTs, giving ChatGPT secure, governed access to your APIs and grounded knowledge.

Gemini & Google AI Studio

Connect Gemini-powered agents to the same MCP server so any Google AI workflow can call your tools through a single standardised interface.

Any MCP-compatible Client

If it speaks the Model Context Protocol, it works. Connect open-source agents, custom CLIs, or in-house orchestrators with zero protocol glue code.

n8n, Zapier & Workflow Tools

Trigger MCP tool calls from automations to push enterprise data into downstream systems — emails, tickets, CRMs, spreadsheets and beyond.

How the connection works

Three steps from zero to a working MCP integration.

1. Configure

Add your APIs, credentials and knowledge bases inside Artificial Wit AI https://ai.artificialwit.com

2. Authorise

Generate an API key from user profile inside Artificial Wit AI https://ai.artificialwit.com

3. Connect

Point any MCP client to https://api.artificialwit.com/mcp and include the x-api-key header using the API key obtained from the user profile.