Hosted MCP server

What it is

The server speaks MCP over streamable HTTP and authenticates with an API key sent as a bearer header (Authorization: Bearer atx_…) — the same key the REST API uses. Tool discovery is keyless; tool calls that cost credits spend from the same balance as the REST API, so create a key and you're set.

Connect a client

Claude Code

Add the server over HTTP with your API key as a header:

claude mcp add --transport http textify https://mcp.textify.me/mcp \
  --header "Authorization: Bearer atx_YOUR_API_KEY"

Claude.ai (custom connector)

1. Open Settings → Connectors → Add custom connector.
2. Name it Textify and paste the URL https://mcp.textify.me/mcp.
3. Authenticate with your API key as an Authorization: Bearer atx_YOUR_API_KEY header. If your client can't attach a header to a remote connector, use the npm shim (stdio) below, which passes the key for you; the tools then appear in chat.

Cursor

Add to ~/.cursor/mcp.json (or a project .cursor/mcp.json):

{
  "mcpServers": {
    "textify": {
      "url": "https://mcp.textify.me/mcp",
      "headers": {
        "Authorization": "Bearer atx_YOUR_API_KEY"
      }
    }
  }
}

Any streamable-HTTP client

Generic MCP client config — point it at the URL and pass your key as a header:

{
  "mcpServers": {
    "textify": {
      "type": "http",
      "url": "https://mcp.textify.me/mcp",
      "headers": {
        "Authorization": "Bearer atx_YOUR_API_KEY"
      }
    }
  }
}

Local config via the npm shim (stdio)

For editors that only speak stdio, the @textifyme/mcp package bridges to the hosted server:

{
  "mcpServers": {
    "textify": {
      "command": "npx",
      "args": ["-y", "@textifyme/mcp"],
      "env": { "TEXTIFY_API_KEY": "atx_YOUR_API_KEY" }
    }
  }
}

Tools

Five tools, each mapping to a REST endpoint and billed the same way. Schemas below are rendered from the server's tool manifest.

{
  "url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ"
}

{
  "videoId": "dQw4w9WgXcQ",
  "title": "Rick Astley - Never Gonna Give You Up (Official Video) (4K Remaster)",
  "language": "en",
  "text": "Never gonna give you up Never gonna let you down"
}

{
  "audio_url": "https://cdn.example.com/interviews/ep-42.mp3",
  "duration_sec": 372
}

{
  "language": "en",
  "durationSec": 372,
  "text": "Welcome back to the show..."
}

{
  "image_url": "https://cdn.example.com/receipts/2026-07-01.jpg"
}

{
  "text": "WHOLE FOODS MARKET\nBananas 1.29\nOat milk 4.49\nTOTAL 5.78",
  "confidence": 0.98
}

{
  "file_url": "https://cdn.example.com/reports/q2-2026.docx"
}

{
  "kind": "docx",
  "format": "markdown",
  "text": "# Q2 2026 Report\n\n## Summary\n..."
}

{
  "url": "https://example.com/blog/shipping-fast"
}

{
  "title": "Shipping Fast Without Breaking Things",
  "wordCount": 812,
  "markdown": "# Shipping Fast Without Breaking Things\n\n..."
}

Agent-flow example: summarize a YouTube video

Once the server is connected, an agent can chain the tools on its own. Ask it to summarize a talk, and it fetches the transcript first, then writes the summary from the returned text — no scraping, no manual copy-paste:

"Summarize this talk in three bullets: https://www.youtube.com/watch?v=dQw4w9WgXcQ"

// The agent calls the MCP tool:
{
  "name": "youtube_transcript",
  "arguments": { "url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ" }
}

// Textify returns the transcript, which the agent then summarizes:
{
  "videoId": "dQw4w9WgXcQ",
  "title": "Rick Astley - Never Gonna Give You Up (Official Video) (4K Remaster)",
  "language": "en",
  "text": "Never gonna give you up Never gonna let you down"
}

You can drive the same flow programmatically from the Claude Messages API using its MCP connector — Textify runs the tool server-side and hands the transcript back to the model to summarize:

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic();

const message = await client.beta.messages.create({
  model: "claude-opus-4-8",
  max_tokens: 1024,
  betas: ["mcp-client-2025-11-20"],
  mcp_servers: [
    {
      type: "url",
      name: "textify",
      url: "https://mcp.textify.me/mcp",
      authorization_token: "atx_YOUR_API_KEY",
    },
  ],
  tools: [{ type: "mcp_toolset", mcp_server_name: "textify" }],
  messages: [
    {
      role: "user",
      content:
        "Fetch the transcript of https://www.youtube.com/watch?v=dQw4w9WgXcQ and summarize it in 3 bullets.",
    },
  ],
});

console.log(message.content);