youtube-transcript-mcp

# YouTube Transcript MCP Server

> Give Claude, ChatGPT, Cursor, and any MCP client the ability to read **YouTube transcripts** — free, no API key, no signup.

A hosted [Model Context Protocol](https://modelcontextprotocol.io) server that exposes one tool, `get_youtube_transcript`, backed by [youtube-transcript.ai](https://youtube-transcript.ai). Paste a YouTube link mid-conversation and your AI reads the full transcript on demand — no copy-paste, no browser tab.

**Connector URL:**

```
https://youtube-transcript.ai/mcp
```

That's it. It's hosted — there's nothing to install or run. Add the URL to your client below.

---

## Why

- **Free & keyless** — no account, no token, no quota wall.
- **Actually reliable** — a hardened fetch pipeline with proxy fallback returns transcripts even on videos that block typical scrapers, so your model rarely comes back empty-handed.
- **Any language** — pick a caption track by language code (`en`, `zh-CN`, `es`, `ja`, …) or let it choose the default.
- **Built for agents** — one clean tool, plain-text output with title and metadata.

## Quick start

### Claude (web / desktop)

Settings → **Connectors** → **Add custom connector**

| Field | Value |
|-------|-------|
| Name  | `youtube-transcript.ai` |
| URL   | `https://youtube-transcript.ai/mcp` |

Then ask: *“Summarize this video: https://youtu.be/dQw4w9WgXcQ”*

### ChatGPT

In the custom connector / MCP server setup, add an MCP server with URL `https://youtube-transcript.ai/mcp` (transport: **Streamable HTTP**), enable it for your chat, then paste a YouTube link.

### Cursor / Windsurf / Cline / Claude Desktop (config file)

For clients that launch MCP servers via a config file, bridge the remote server with [`mcp-remote`](https://www.npmjs.com/package/mcp-remote):

```json
{
  "mcpServers": {
    "youtube-transcript": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://youtube-transcript.ai/mcp"]
    }
  }
}
```

Clients that support remote MCP URLs natively can use `https://youtube-transcript.ai/mcp` directly.

## Tool reference

### `get_youtube_transcript`

Fetch the full text transcript of a YouTube video.

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `video`   | string | yes | YouTube URL (watch, `youtu.be`, shorts, embed) or a bare 11-character video ID |
| `lang`    | string | no  | BCP-47 language code, e.g. `en`, `zh-CN`, `es`. Omit for the default track |

Returns plain text: the transcript with title, duration, and a list of available languages.

## Example prompts

- “Summarize the key points of `<link>` in 5 bullets.”
- “Pull the transcript of this video and translate it to Spanish.”
- “What does the speaker say about pricing in `<link>`? Quote the exact lines.”

## Programmatic use

The server speaks standard MCP over Streamable HTTP (JSON-RPC). A zero-dependency example is in [`examples/quickstart.mjs`](examples/quickstart.mjs):

```bash
node examples/quickstart.mjs https://youtu.be/dQw4w9WgXcQ
```

## Notes

- **Fair use:** the endpoint is rate-limited per IP to keep it free for everyone. For heavy/automated workloads, get in touch.
- **Not affiliated** with YouTube or Google. Transcripts are fetched from publicly available captions.

## Links

- Web tool (no AI client needed): https://youtube-transcript.ai
- Connector landing page: https://youtube-transcript.ai/mcp
- Ready-made transcripts: https://youtube-transcript.ai/docs

## License

MIT