TranscriptFetch MCP Server
A Model Context Protocol server that gives any MCP client (Claude Desktop, Cursor, and others) access to the TranscriptFetch API: a production transcript API for YouTube, TikTok and Instagram, with AI transcription when captions are missing. Fetch transcripts, search videos, list channels and playlists, and check your credit balance.
Get an API key
- Create a free account at transcriptfetch.com. No card needed.
- Open Dashboard, then API keys and create a key. It starts with
tf_live_.
- Put it in
TRANSCRIPTFETCH_API_KEY in the client configuration below.
Every account gets 50 free credits a month. Failures are free: a fetch that returns no transcript is never charged.
Runs locally over stdio and calls the TranscriptFetch API with your key. Prefer a hosted, remote server? Point your client at https://transcriptfetch.com/mcp instead (OAuth or API key). The hosted server waits inline for short-form AI transcription, so no polling is needed there.
| Tool | What it does |
|---|
get_transcript | Transcript for a video. YouTube, TikTok, Instagram, or a direct media URL. Set ai_fallback: true to transcribe the audio when no captions exist (typically ~30 seconds for short videos, longer for long ones) |
search_videos | Search YouTube by keyword |
list_channel_videos | List a YouTube channel's videos (handle, ID, or URL) |
list_playlist_videos | List a YouTube playlist's videos (ID or URL) |
get_credits | Remaining credit balance for the key. Never billed |
Pricing is per successful result: a caption fetch or a video list costs 1 credit, and AI transcription of the audio costs 1 credit per started minute of audio, charged only on delivery. Failed, blocked and empty results are never charged, which matters on short-form video where many clips have no speech at all.
Install
No install needed. Run it on demand with npx:
TRANSCRIPTFETCH_API_KEY=tf_live_... npx -y transcriptfetch-mcp
Or install globally:
npm install -g transcriptfetch-mcp
Requires Node 18+.
Run from source
git clone https://github.com/TranscriptFetch/mcp-server
cd mcp-server && npm install && npm run build
Then point your client at the built entrypoint with "command": "node" and
"args": ["/absolute/path/to/mcp-server/dist/index.js"].
Client configuration
Claude Desktop
Add this to claude_desktop_config.json (Settings then Developer then Edit Config):
{
"mcpServers": {
"transcriptfetch": {
"command": "npx",
"args": ["-y", "transcriptfetch-mcp"],
"env": { "TRANSCRIPTFETCH_API_KEY": "tf_live_..." }
}
}
}
Cursor
Add the same block under mcpServers in your Cursor MCP settings.
Restart the client, and the five tools appear.
Example
Once connected, ask your assistant naturally:
Get the transcript for https://youtu.be/aircAruvnKk and summarize the key points.
Search YouTube for "how transformers work" and list the top 5 videos.
List the latest videos from @lexfridman and pull the transcript of the newest one.
How many TranscriptFetch credits do I have left?
The assistant picks the matching tool and works from the returned transcript or video list.
Configuration
| Env var | Required | Default |
|---|
TRANSCRIPTFETCH_API_KEY | yes | none |
TRANSCRIPTFETCH_BASE_URL | no | https://transcriptfetch.com |
Docker
The server speaks MCP over stdio, so there is no port to expose. -i is
required: without an attached stdin the transport closes immediately and the
container looks like it crashed.
docker build -t transcriptfetch-mcp .
docker run --rm -i -e TRANSCRIPTFETCH_API_KEY=tf_live_... transcriptfetch-mcp
Links
License
MIT