目录 / youtube-transcripts
youtube-transcripts
## What it does Gives any MCP-compatible agent (Claude, Cursor, VS Code, ChatGPT, Codex) clean YouTube transcripts as timestamped markdown, ready to use as context. No scraping, no proxies, no YouTube rate limits to manage: the server handles the fetch and caches every result. ## Tools | Tool | What it does | Cost | |---|---|---| | `get_transcript` | Full transcript of one video (ID or URL) | 1 credit | | `get_channel_transcripts` | Recent videos from a channel, newest first (limit 1-50) | 1 credit per video | | `get_playlist_transcripts` | Videos from a playlist, in order (limit 1-50) | 1 credit per video | | `polish_transcript` | Cleaned auto-captions: punctuation restored, filler removed, faithful to what was said | ~3 credits per 1,000 words (min 5) | All tools are read-only. Videos without captions fail free. Rate limit: 5 requests per 10 seconds. ## Setup Get a free API key (25 free credits, no card) at [youtubetranscriptdownload.com/account](https://youtubetranscriptdownload.com/account), then add the server: ```json { "mcpServers": { "ytd": { "url": "https://youtubetranscriptdownload.com/api/mcp", "headers": { "Authorization": "Bearer ytd_sk_YOUR_KEY" } } } } ``` Claude Code: ```bash claude mcp add --transport http \ --header "Authorization: Bearer ytd_sk_YOUR_KEY" \ ytd https://youtubetranscriptdownload.com/api/mcp ``` ## Example output `get_transcript` returns the video title, source link, timestamped text, and a credit footer: ```text How Transformers Work Source: https://youtube.com/watch?v=... [00:00:12] The transformer architecture was introduced in the 2017 paper Attention Is All You Need by Vaswani et al. ... --- 1 credit used · 42 remaining ``` `get_channel_transcripts` and `get_playlist_transcripts` return one section per video, separated by dividers; videos without captions are noted and cost nothing: ```text Lex Fridman Podcast · 3 videos [transcript of video 1] --- [transcript of video 2] --- abc123xyz: No transcript available (skipped, not charged) --- 2 credits used · 40 remaining ``` `polish_transcript` turns raw auto-captions into clean text, faithful to what was said: ```text Before: so um the transformer architecture was introduced uh in the attention is all you need paper in in 2017 After: The transformer architecture was introduced in the Attention Is All You Need paper in 2017. ``` ## Links - Setup guide for all clients: https://youtubetranscriptdownload.com/mcp - REST API docs (same backend): https://youtubetranscriptdownload.com/youtube-transcript-api - Pricing: https://youtubetranscriptdownload.com/pricing
接入信息
- 传输形态
- http
- 鉴权方式
- 需 API Key(需要配置:apiKey)
- 端点
https://youtube-transcripts--hello-ympt.run.tools
{
"mcpServers": {
"youtube-transcripts": {
"headers": {
"Authorization": "Bearer \u003cYOUR_KEY\u003e"
},
"url": "https://youtube-transcripts--hello-ympt.run.tools"
}
}
}
能力清单
| 工具 | 说明 |
|---|---|
| get_transcript | Get the full transcript of a single YouTube video as timestamped markdown. Read-only: fetches existing captions, modifies nothing. Requires an API key; each successful call charges 1 credit, including repeat calls for the same video, so reuse a transcript already in context instead of re-fetching. Videos without captions return an error and cost nothing. Rate limit: 5 requests per 10 seconds. |
| polish_transcript | Get a cleaned-up transcript of a YouTube video's auto-generated captions: punctuation and capitalisation restored, filler and false starts removed, paragraphs added, misheard names fixed, faithful to what was said. Use when raw captions are too messy to read or quote; for a plain transcript use get_transcript. Read-only; requires an API key. Each call charges credits by transcript length (about 3 per 1,000 words, minimum 5), including repeat calls, so keep the result in context. Human-uploaded captions (already clean) and transcripts over ~7,000 words return an error without charging. Rate limit: 5 requests per 10 seconds. |
| get_channel_transcripts | Get transcripts for a YouTube channel's most recent videos (newest first) as timestamped markdown, one section per video. Use for research across a creator's recent output; for one known video use get_transcript. Read-only; requires an API key. Charges 1 credit per video that returns a transcript, including repeat calls; videos without captions are skipped free. A 10-video call typically costs up to 10 credits, so start with a small limit. Rate limit: 5 requests per 10 seconds. |
| get_playlist_transcripts | Get transcripts for the videos in a YouTube playlist (in playlist order) as timestamped markdown, one section per video. Use for working through a course, series, or curated list; for one known video use get_transcript. Read-only; requires an API key. Charges 1 credit per video that returns a transcript, including repeat calls; videos without captions are skipped free. A 10-video call typically costs up to 10 credits, so start with a small limit. Rate limit: 5 requests per 10 seconds. |
提交举报 / 纠错
侵权举报经核验成立后,我们会即时下线该条目并删除已存的内容副本。