YouTube Transcript AI

ai.youtube-transcriptPublished by youtube-transcript.aiOfficial publisherv1.3.0Updated Oct 29, 2026

Read YouTube transcripts and search what people say across 425k videos. Free API key.

VerifiedStreamable HTTPWeb executableWeb Search & ScrapingData & AnalyticsMedia & Design

Overview

AI-generated overview

Hosted MCP server that lets an assistant read YouTube transcripts and search what people say across a large library of videos.

What you can do
  • Fetch full transcripts of a video by URL or ID, with language selection and paging for long ones
  • Search YouTube videos and list a channel's or playlist's latest videos
  • Find keywords inside a transcript with timestamps and jump links, or read several transcripts in one batch
  • Search spoken sentences across a large video library and read the surrounding context
  • Rank videos and channels by how often they mention a topic, and chart mentions over time
  • Fetch a video's top comments
When to use it
Useful when you want an assistant to summarize, quote or fact-check YouTube videos, research a topic across many creators, or track how often something is said over time, without installing anything locally.
Requirements
Remote Streamable HTTP endpoint; no local install. Requires an API key sent as an Authorization Bearer header or as a key query parameter in the connector URL. A free key is created with a Google sign-in. Videos need captions. Free tier allows 10 requests a day; Pro is 1,000 a day for $120 a year.
Before you install
The API key is a credential: in the URL form it sits in the connector URL, so treat that URL as secret. Every tool call, including each part of a long transcript, counts against the daily quota, and the paid tier costs money. Requests are rate limited to about 10 per 10 seconds. Transcripts and comments come from public captions and are not affiliated with YouTube or Google.

Installation

In SourceWeft

  1. Open YouTube Transcript AI in the dashboard and add it to a workspace.
  2. Connect Authorization token credentials. They are stored privately and never shown on public pages.
  3. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/ai.youtube-transcript%2Fyoutube-transcript/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "youtube-transcript-ai": {
      "type": "http",
      "url": "https://youtube-transcript.ai/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_TOKEN>"
      }
    }
  }
}

README

YouTube Transcript MCP Server

Give Claude, ChatGPT, Cursor, Claude Code and any MCP client the ability to read YouTube transcripts. Hosted, nothing to install. Free API key: 10 requests a day.

A hosted Model Context Protocol server that exposes one tool, get_youtube_transcript, backed by youtube-transcript.ai. Paste a YouTube link mid-conversation and your AI reads the full transcript on demand.

Connector URL:

https://youtube-transcript.ai/mcp?key=YOUR_KEY

Replace YOUR_KEY with your API key. Create a free key with a Google sign-in; it takes about a minute.

Why

  • Free to start. A free key covers 10 transcripts a day. Pro is 1,000 a day for $120 a year, on the same key.
  • Reliable. A hardened fetch pipeline with proxy fallback returns transcripts even on videos that block typical scrapers.
  • Any language. Pick a caption track by language code (en, zh-CN, es, ja, …) or let it choose the default.
  • Built for agents. One tool, plain-text output with title and metadata. Multi-hour videos come back in parts, so nothing gets cut off.

Authentication

Tool calls need an API key. There is no OAuth. Send the key in one of two ways:

WhereExample
HeaderAuthorization: Bearer YOUR_KEY
URLhttps://youtube-transcript.ai/mcp?key=YOUR_KEY

The URL form is for clients that only accept a connector URL.

Quick start

Claude (web / desktop)

Settings → Connectors → Add custom connector

FieldValue
Nameyoutube-transcript.ai
URLhttps://youtube-transcript.ai/mcp?key=YOUR_KEY

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?key=YOUR_KEY (transport: Streamable HTTP), enable it for your chat, then paste a YouTube link.

Claude Code

bash
claude mcp add --transport http --scope user youtube-transcript \  https://youtube-transcript.ai/mcp \  --header "Authorization: Bearer YOUR_KEY"

Cursor

Add to ~/.cursor/mcp.json, or to .cursor/mcp.json inside one project:

json
{  "mcpServers": {    "youtube-transcript": {      "url": "https://youtube-transcript.ai/mcp",      "headers": { "Authorization": "Bearer YOUR_KEY" }    }  }}

Codex

Add to ~/.codex/config.toml:

toml
[mcp_servers.youtube-transcript]url = "https://youtube-transcript.ai/mcp"http_headers = { "Authorization" = "Bearer YOUR_KEY" }

Clients that only launch local servers

For clients that start MCP servers from a config file and cannot take a remote URL, bridge with mcp-remote:

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

Let your agent do it

Paste https://youtube-transcript.ai/mcp.md into a coding agent. It is a plain-text version of these instructions, written for an AI to follow.

Tool reference

get_youtube_transcript

Fetch the full text transcript of a YouTube video.

ParameterTypeRequiredDescription
videostringyesYouTube URL (watch, youtu.be, shorts, embed) or a bare 11-character video ID
langstringnoCaption language code, e.g. en, zh-CN, es. Omit for the default track
offsetintegernoWhere to continue a long transcript. Use the value given at the end of the previous part

Returns plain text: the transcript with title, duration, and a list of available languages. Transcripts over 120,000 characters come back in parts; each part ends with a note giving the offset for the next call.

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:

bash
YT_TRANSCRIPT_KEY=YOUR_KEY node examples/quickstart.mjs https://youtu.be/dQw4w9WgXcQ

Calling from your own code without MCP? There is a plain HTTP endpoint too: see the YouTube Transcript API.

Limits

  • Free: 10 requests a day. Pro: 1,000 a day for $120 a year. Upgrade from the key page.
  • Every tool call counts as one request, including each part of a long transcript. The count resets at 00:00 UTC.
  • The video needs captions, either written by the uploader or auto-generated by YouTube.

Notes

  • Not affiliated with YouTube or Google. Transcripts are fetched from publicly available captions.
  • Listed in the official MCP Registry as ai.youtube-transcript/youtube-transcript.

Links

License

MIT

Source: README.md at commit 5c71e9f

Tools

11

get_youtube_transcript

Fetch the full text transcript of a YouTube video. Accepts a YouTube URL (watch, youtu.be, shorts, embed) or a bare 11-character video ID, and returns the transcript as plain text with title and metadata. Optionally pass a BCP-47 language code to select a specific caption track. Very long transcripts (multi-hour videos) come back in parts of up to 120,000 characters: when there is more, the result ends with a note giving the offset to pass on the next call.

Parameterslangvideorequiredoffset

search_youtube_videos

Search YouTube and return matching videos with channel, view count and duration. Use it first when researching a topic: run several differently-worded queries (and in more than one language), then read the best hits with get_youtube_transcript, find_in_transcript or get_transcripts_batch. Optional filters on views and length.

Parameterssortlimitqueryrequiredmin_viewsmax_minutesmin_minutes

find_in_transcript

Find where a video talks about something. Give one or more keywords and get back only the matching passages, each with a timestamp and a link that opens the video at that moment. Far cheaper than reading the whole transcript when you are looking for specific points. Matching is a case-insensitive text match, so try synonyms and spelling variants (auto-captions often mis-hear names).

Parameterslangvideorequiredcontextkeywordsrequiredmax_hits

get_transcripts_batch

Read the transcripts of several videos in one call (up to 5), each cut to a length you choose, so you can skim a set of search results without filling your context. Each video counts as one request against your daily quota. For the rest of a long video use find_in_transcript or get_youtube_transcript.

Parameterslangvideosrequiredmax_chars_each

get_video_comments

Fetch the top comments of a YouTube video (YouTube's "top" ordering, replies not included), with like counts. Comments are where viewers report what worked for them, which makes them useful for checking claims made in the video.

Parameterslimitvideorequired

list_channel_videos

List the latest videos of a YouTube channel or playlist, with view count and duration. Use it to follow a creator who looks credible, or to cover a playlist.

Parameterslimitchannelrequired

search_spoken_words

Search what people actually say inside YouTube videos. Full-text search over about 100 million spoken sentences from about 425,000 English-language videos (the ones users looked up on youtube-transcript.ai), with typically 1 to 8 million new sentences added every day. Returns matching sentences, each with the video, channel and a link that opens the video at that moment. Use it to find evidence and quotes, to see how people talk about a topic, product or phrase, or to find which videos mention something. Captions are auto-generated and often mis-hear names (for example "Claude Code" is often written "cloud code"), so pass likely mis-hearings in also_match. The library is a sample of YouTube, not all of it. Dates are the day we captured the captions, not the day the video was published.

Parameterslimitqueryrequiredsinceuntilchannelalso_match

get_spoken_context

Read what was said around a moment in a video: the sentences before and after a timestamp. Use it after search_spoken_words to see what a matching sentence was part of, before you quote or rely on it. Works for videos in the library only.

Parametersvideorequiredwindowsecondsrequired

rank_videos_by_topic

Find the videos that talk about something the most, ranked by how many sentences mention it. Better than searching titles when you want the videos that really go deep on a topic. Each result has the number of mentions, the first mention and a link. Read the best ones with get_youtube_transcript or find_in_transcript. Same library, caption quirks and dates as search_spoken_words.

Parameterslimitqueryrequiredsinceuntilchannelalso_match

rank_channels_by_topic

Find which channels talk about something the most, ranked by mentions, with how many of their videos mention it. Use it to find who covers a topic: creators to follow, experts, or partners. Same library, caption quirks and dates as search_spoken_words.

Parameterslimitqueryrequiredsinceuntilalso_match

spoken_mentions_over_time

Count how often something is mentioned over time, by day, week or month. Each period shows the mentions and the rate per million spoken sentences, so busy and quiet periods compare fairly. Use it to see whether a topic is rising or fading. The date is when we captured the captions, not when the video was published, and the first day (2026-09-13) holds a one-off backfill of earlier captions, so compare the rate, not the raw count, and be careful with that first period.

Parametersqueryrequiredsinceuntilbucketchannelalso_match

Version history

3
  1. v1.3.0LatestOct 29, 2026
  2. v1.1.0Oct 14, 2026
  3. v1.0.0Sep 16, 2026

Score

100% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listedRequired15 pts
  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts