500Toolbox MCP

npx add-mcp 'https://toolkitbackend-j69m.onrender.com/api/v1/mcp/connect'

30 tools

100 free credits / month

Nothing to install

The standard way for AI models to use tools

The Model Context Protocol is an open standard that lets an AI assistant call external tools. Without it a model can only tell you how to check a domain’s mail records. With it, the model checks them and answers from the real result.

Our server publishes every endpoint as an MCP tool with its input schema and credit cost, so the assistant knows how to call each one and what it will spend.

You ask

The model picks a tool

We run it

It answers from data

Add it to your AI app in one click

Paste one URL into Claude or ChatGPT, or use the one-click button for Cursor and VS Code. You sign in once - there is no API key to copy and no file to edit.

Server URL https://toolkitbackend-j69m.onrender.com/api/v1/mcp/connect

Fastest way

1. Click "Open Claude connectors" below - claude.ai opens on Settings → Connectors.
2. Click "Add custom connector". Name: 500Toolbox. URL: use the Copy button above and paste it.
3. Click Add, then Connect. Sign in to 500Toolbox and click Allow. Done.
4. In a chat, open the tools menu (the + / sliders icon) and switch 500Toolbox on.

A connector added on claude.ai also appears in the Claude desktop and mobile apps. On Team and Enterprise plans an owner adds it once under Organization settings → Connectors; members then click Connect.

Advanced: connect with an API key (config file or stdio bridge) For clients without sign-in support, or scripts that should use a dedicated key.

1. Open Cursor Settings → Tools & MCP → New MCP server (this opens mcp.json).
2. Paste the block, put your key in the header and save.
3. The server turns green and its tools appear in Agent mode.

{
  "mcpServers": {
    "toolbox": {
      "url": "https://toolkitbackend-j69m.onrender.com/api/v1/mcp",
      "headers": {
        "Authorization": "Bearer tb_live_your_key"
      }
    }
  }
}

Every config below uses the placeholder tb_live_your_key. Create a real key on the API keys page.

Want your Google Search Console data in Claude or ChatGPT? That is a separate connector that signs in with one click and needs no key. See the Search Console MCP →

What your assistant gets

Every tool appears automatically once connected. Prices are per call and shown to the model, so it prefers a cheap lookup over an expensive crawl.

Network & DNS

Web & HTTP

SEO & Content

[

seo-meta 2 credits

](https://www.500toolbox.com/api/endpoints/seo-meta)

Hashing & Passwords

Data & Utility

Documents

Images

Things people ask once it is connected

“What are the MX records for example.com, and is its SSL certificate expiring soon?”

“Convert this Word document to PDF and give me the file.”

“Remove the background from this product photo.”

“Check every link on this page and list the broken ones.”

“Take a full-page screenshot of our pricing page on mobile.”

“Which country is this IP in, and is it a hosting provider?”

Fair, predictable, visible

Free monthly credits

Pay only for success

Prices reflect real work

Credits never expire

Questions about the MCP server

Do I need to install anything?

Not for HTTP clients like Cursor, VS Code, Windsurf or Claude Code - they connect to the hosted URL directly. Claude Desktop and other stdio-only clients run a tiny bridge through npx (@500toolbox/mcp), which needs Node.js 18 or newer.

Can I see the tools before I have a key?

Yes. Listing the tools is free and needs no key, so a client connects and shows the whole catalogue. A key is only required to run a tool.

How does the model know what a tool costs?

Each tool description includes its credit price, and the server tells the model to prefer the cheaper tool when either would answer the question.

What happens when I run out of credits?

The call is refused with an "insufficient_credits" message the assistant reads back to you. Nothing is charged. Top up on the billing page or wait for the monthly free credits.