Connect Your AI

You probably already work with an AI assistant. Connect it to the Prometora docs and it can answer questions about what the platform supports, how to configure it, and how features work, with reliable answers pulled straight from the documentation.

What's available

• Docs MCP server - https://www.prometora.com/api/mcp/docs. Gives your AI three tools: search_docs (find relevant pages), get_page (read a full page as markdown), and list_pages (overview of every page).
• llms.txt - www.prometora.com/llms.txt. An index of every docs page with a one-line description and a markdown link, following the llms.txt convention.
• Markdown mirrors - append .md to any docs URL. For example /docs/store-settings/shipping.md is the shipping page as clean markdown, much easier for an AI to read than the HTML.
• llms-full.txt - the entire documentation in a single file for tools that prefer one big context load.

Connect your AI via MCP

The Model Context Protocol (MCP) is the standard way to give AI tools access to external knowledge. The Prometora docs server uses streamable HTTP with no authentication, so setup is one step in every client.

Claude Desktop and claude.ai

1. Open Settings → Connectors and click Add custom connector.
2. Name it Prometora Docs and paste this URL (no authentication needed):

https://www.prometora.com/api/mcp/docs

That's it. Now just chat: ask something like “How does buyer approval work on Prometora?” and Claude searches the docs and answers from them, citing the pages it used.

Claude Code

Run this once in your terminal:

claude mcp add --transport http prometora-docs https://www.prometora.com/api/mcp/docs

Then start a new Claude Code session. You can check the connection with the /mcp command (it should list prometora-docs as connected), and from there you simply ask questions in the chat:

How do sellers get paid on a Prometora marketplace?

Claude decides on its own when to use the docs tools, so there is nothing else to invoke. If you ever want to point it at the docs explicitly, mention the server by name: “Using prometora-docs, explain how shipping deadlines work.”

ChatGPT

1. Open Settings → Connectors (on a paid plan, enable Developer mode under Advanced if you don't see the option to add one).
2. Add a new connector named Prometora Docs with this MCP server URL and no authentication:

https://www.prometora.com/api/mcp/docs

Then enable the connector in a chat and ask your question. If your ChatGPT plan doesn't offer connectors, use the llms.txt method below instead; it works in any plan.

Cursor

Add this to ~/.cursor/mcp.json (or the project's .cursor/mcp.json):

{
  "mcpServers": {
    "prometora-docs": {
      "url": "https://www.prometora.com/api/mcp/docs"
    }
  }
}

VS Code

Add this to .vscode/mcp.json in your workspace:

{
  "servers": {
    "prometora-docs": {
      "type": "http",
      "url": "https://www.prometora.com/api/mcp/docs"
    }
  }
}

Other MCP clients

Windsurf and any other MCP-capable tool: add a remote server with transport streamable HTTP, URL https://www.prometora.com/api/mcp/docs, and no authentication.

No MCP? Point your AI at the markdown

Any AI that can fetch a URL can use the docs without any setup. For example, paste this into ChatGPT, Claude, or any assistant with web access:

Read https://www.prometora.com/llms.txt, then answer:
how do sellers get paid on a Prometora marketplace?

The assistant will find the right page in the index and read its markdown version. This is also what AI tools that support the llms.txt convention do automatically.

What it's good for

• “Does Prometora support per-listing buyer approval, and how do I enable it?”
• “What's the difference between self-serve and managed seller onboarding?”
• “Which webhook events exist and what payload do they send?”
• “Walk me through connecting a custom domain, step by step.”
• Evaluating Prometora: let your AI read the docs and tell you whether the platform fits your marketplace idea.