# YouTube MCP

> YouTube search: find videos, fetch details and comments, pull full transcripts.

- name: ai.wondel/youtube
- version: 1.0.0
- connection class: R0 (autonomous)
- trust: 83/100 flags: no-repository
- quality: 84/100 (new)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/ai.wondel%2Fyoutube
- tools: https://api.protogrid.dev/v1/servers/ai.wondel%2Fyoutube/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "youtube": {
      "type": "http",
      "url": "https://mcp.wondel.ai/youtube/mcp"
    }
  }
}
```


## Trust

- hygiene: 80
- liveness: 100
- freshness: 100
- provenance: 55
- drivers: +dns-namespace +website +reachable +uptime-100% +updated-5d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 84/100 (new)

### protocol: 100 (weight 25)

- pass `protocol.modern`: supports 2026-07-28 (server/discover)
- pass `protocol.stateless`: answers without a session
- pass `protocol.transport`: streamable HTTP
- pass `protocol.list_ttl`: tool list is cacheable (ttlMs 3600000)

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 71 (weight 30)

- pass `tools.descriptions`: every tool has a description
- pass `tools.description_length`: descriptions are concise
- pass `tools.schemas`: every tool has an object input schema
- fail `tools.annotations`: no tool declares readOnlyHint or destructiveHint, so clients cannot tell safe calls from risky ones
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: youtube_search (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); youtube_video (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); youtube_transcript (readOnlyHint, destructiveHint, idempotentHint, openWorldHint)
- pass `tools.token_cost`: about 943 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: n/a (weight 15)

- n/a `stability.changes`, `stability.rug_pull`: tool history recorded for 5 of the 7 days needed

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 3, about 943 tokens to load them all
- tool set hash: 7a076a4ffc2054fca45dab022d09f390
- badge: [![protogrid quality](https://protogrid.dev/badge/ai.wondel/youtube.svg)](https://protogrid.dev/servers/ai.wondel/youtube)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

_No tool definition changes recorded._

## Remotes

- https://mcp.wondel.ai/youtube/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

_none_

## Tools (3)

- `youtube_search`: Search YouTube live via SerpAPI. Returns video_results (videos), plus channels, playlists, movies and related searches. Each video result's id is the YouTube video id — feed it as `v` to youtube_video (details, related videos, comments) and to youtube_transcript. `sp` carries SerpAPI filter and next-page tokens from a prior result for pagination / refined filtering.
- `youtube_transcript`: Fetch a YouTube video's transcript via SerpAPI. `v` is the video id from a youtube_search result. Pick a track with language_code (default 'en' — use the video's own language), and optionally title and type ('asr' for auto-generated captions). When the requested track doesn't exist the tool falls back once to an available track (same language first, manual before auto-generated) and says so in `note`.
- `youtube_video`: Fetch a single YouTube video by id via SerpAPI: details, related videos, comments and replies. `v` is the video id from a youtube_search video_results entry. Page related videos / comments / replies with next_page_token from a prior result.