# kirbah-mcp-youtube

MCP server that delivers structured YouTube transcripts and statistics without payload bloat. It runs transcript lookups without keys and cuts context tokens up to 87%.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 32
- Author: kirbah
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/kirbah/mcp-youtube
- Updated: 2026-10-20
- HTML: https://affpapa.org/en/nevibecode/mcp/kirbah-mcp-youtube

## What it does

Standard YouTube API responses carry redundant thumbnails, localization details, and nested eTags. This server strips metadata before passing data to the LLM. Channel statistics shrink from 1.9 KB down to 0.2 KB. The call consumes 86 tokens instead of 673.

Captions work right away without configuration. Point the agent at `youtube://transcript/{videoId}` without credentials. Deep search and channel metrics require the `YOUTUBE_API_KEY` variable.

YouTube allows only 10,000 daily quota units. Looping agents drain that limit fast. Passing a MongoDB connection string in `MDB_MCP_CONNECTION_STRING` avoids burning points on repeated lookups.

## Requirements
- Remote server: `https://server.smithery.ai/@kirbah/mcp-youtube/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http kirbah-mcp-youtube https://server.smithery.ai/@kirbah/mcp-youtube/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add kirbah-mcp-youtube --url https://server.smithery.ai/@kirbah/mcp-youtube/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "kirbah-mcp-youtube": {
            "url": "https://server.smithery.ai/@kirbah/mcp-youtube/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "kirbah-mcp-youtube": {
            "type": "http",
            "url": "https://server.smithery.ai/@kirbah/mcp-youtube/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http kirbah-mcp-youtube https://server.smithery.ai/@kirbah/mcp-youtube/mcp
```

## When to use it
- extract video captions for LLM summarization without API keys
- trim token consumption when requesting YouTube channel metrics
- query video search results without nested thumbnail metadata
- prevent YouTube API daily quota exhaustion using MongoDB cache

## Example prompts
- Read the transcript for `youtube://transcript/dQw4w9WgXcQ` and summarize main points.
- Get channel stats and subscriber count for @veritasium.
- Search for trending videos about rust programming and show their view counts.

## Similar
- [REA](https://affpapa.org/en/nevibecode/mcp/rea.md) — MCP server · ★ 386989
- [GitHub](https://affpapa.org/en/nevibecode/mcp/github.md) — MCP server · ★ 33918
- [Netdata](https://affpapa.org/en/nevibecode/mcp/netdata.md) — MCP server · ★ 81115
- [browser-use](https://affpapa.org/en/nevibecode/mcp/browser-use.md) — MCP server · ★ 119160
- [Codebase Memory](https://affpapa.org/en/nevibecode/mcp/codebase-memory.md) — MCP server · ★ 47392
- [mcp-server](https://affpapa.org/en/nevibecode/mcp/mcp-server.md) — MCP server · ★ 43872

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md