YouTube Video Summarizer (pay-per-call)
Summarize YouTube videos or fetch transcripts. Pay-per-call via x402 (USDC on Base), no account.
Should I use this
Quality & Safety
Based on automated analysis of tool definitions and protocol compliance.
Context Cost
This is the approximate number of tokens consumed each time the server's tools are loaded into a model's context. Higher counts reduce the attention available for other tasks.
Install
One-Click Install
Add this to your `claude_desktop_config.json` file:
{
"mcpServers": {
"summarize-vid": {
"url": "https://mcp.sumvid.app/mcp"
}
}
}Remote endpoints
https://mcp.sumvid.app/mcpstreamable-httpWhat it can do
Tool inventory
Tools (4)
⚪summarize_youtube_video(url, language)
Summarize a YouTube video with Claude and return the summary text, in any of 100+ output languages (`language` param, BCP-47). Pay-per-call via x402 (USDC on Base) — no account or API key. Cheaper than pulling the full transcript into your own context. A bad URL or a video without captions returns an error and is not charged. New videos up to 60 minutes; longer ones (or ones whose length can't be determined) return an error and are not charged.
Input Schema
{
"type": "object",
"properties": {
"url": {
"title": "Url",
"type": "string"
},
"language": {
"default": "en",
"title": "Language",
"type": "string"
}
},
"required": [
"url"
],
"title": "summarize_youtube_videoArguments"
}🟢get_youtube_transcript(url, language)
Fetch the raw transcript text of a YouTube video (no model run — the cheaper SKU), in any of 100+ languages (`language` param, BCP-47). Pay-per-call via x402 (USDC on Base), no account. A video without captions returns an error and is not charged.
Input Schema
{
"type": "object",
"properties": {
"url": {
"title": "Url",
"type": "string"
},
"language": {
"default": "en",
"title": "Language",
"type": "string"
}
},
"required": [
"url"
],
"title": "get_youtube_transcriptArguments"
}🟢list_new_channel_videos(channel, since_video_id, since)
New videos on a YouTube channel since your last poll — stateless channel monitoring, no accounts or webhooks. Pass a channel URL, @handle, or channel id, plus an optional cursor (`since_video_id` or `since`, an ISO 8601 timezone-aware timestamp) from your last call. Returns new video ids/titles/publish times, newest first, plus `newest_video_id` to use as your next cursor. Pay-per-call via x402 (USDC on Base), no account or API key. An unresolvable channel or a naive `since` timestamp returns an error and is not charged.
Input Schema
{
"type": "object",
"properties": {
"channel": {
"title": "Channel",
"type": "string"
},
"since_video_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Since Video Id"
},
"since": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Since"
}
},
"required": [
"channel"
],
"title": "list_new_channel_videosArguments"
}⚪format_json(text, pretty, sort_keys)
Validate raw JSON and return it pretty-printed or minified. `pretty` (default true) uses two-space indentation; false minifies. `sort_keys` (default false) sorts object keys alphabetically and leaves array order alone. Input is capped at 1 MB. Invalid JSON raises with the line and column and is not charged. Pay-per-call via x402 (USDC on Base), no account or API key.
Input Schema
{
"type": "object",
"properties": {
"text": {
"title": "Text",
"type": "string"
},
"pretty": {
"default": true,
"title": "Pretty",
"type": "boolean"
},
"sort_keys": {
"default": false,
"title": "Sort Keys",
"type": "boolean"
}
},
"required": [
"text"
],
"title": "format_jsonArguments"
}Community
Evidence