mcpserver-justtranscribe
OfficialAllows transcribing public Facebook videos into accurate, timestamped text, with options to export as SRT, VTT, TXT, Markdown, or CSV.
Allows transcribing public video or audio files hosted on Google Drive into accurate, timestamped text, with options to export as SRT, VTT, TXT, Markdown, or CSV.
Allows transcribing public Instagram videos into accurate, timestamped text, with options to export as SRT, VTT, TXT, Markdown, or CSV.
Allows transcribing public Pinterest videos into accurate, timestamped text, with options to export as SRT, VTT, TXT, Markdown, or CSV.
Allows transcribing public TikTok videos into accurate, timestamped text, with options to export as SRT, VTT, TXT, Markdown, or CSV.
Allows transcribing public YouTube videos into accurate, timestamped text, with options to export as SRT, VTT, TXT, Markdown, or CSV.
justtranscribe-mcp
MCP server for JustTranscribe — transcribe public videos and audio into accurate, timestamped text without leaving your AI assistant. Paste a YouTube, TikTok or Instagram link in the chat; get the transcript back, export SRT/TXT/Markdown, or pull the AI breakdown of a social video. Spanish, English and dozens of languages are auto-detected.
Two ways to connect — both use the same API key and the same account limits:
Transport | For | |
Hosted | Streamable HTTP | Claude Code, Cursor, any client that supports remote servers with headers |
This package | stdio | Claude Desktop and other stdio-only clients |
Setup
Create a free API key at https://justtranscribe.ai/api-keys (free during the beta, no card).
Add the server to your client.
Claude Desktop (claude_desktop_config.json):
{
"mcpServers": {
"justtranscribe": {
"command": "npx",
"args": ["-y", "justtranscribe-mcp"],
"env": { "JUSTTRANSCRIBE_API_KEY": "jt_live_..." }
}
}
}Claude Code (hosted variant — no install):
claude mcp add --transport http justtranscribe https://justtranscribe.ai/api/mcp \
--header "Authorization: Bearer jt_live_..."Related MCP server: faster-whisper-mcp
Claude Code plugin
This repo is also a Claude Code plugin. It runs this package as a local
MCP server and adds a transcribe skill that knows the limits and the
export formats.
/plugin marketplace add anthropics/claude-plugins-community
/plugin install justtranscribe@claude-communitySet your key once (create one free at https://justtranscribe.ai/api-keys) and restart Claude Code:
export JUSTTRANSCRIBE_API_KEY=jt_live_...Then:
/justtranscribe:transcribe https://www.youtube.com/watch?v=jNQXAC9IVRwPrefer signing in to pasting a key? Use the hosted server directly instead
of the plugin — claude mcp add --transport http justtranscribe https://justtranscribe.ai/api/mcp — which runs an OAuth sign-in. (A
plugin's .mcp.json only loads local stdio servers, so the plugin ships
the npx variant.)
To try the plugin before it is listed, clone this repo and run
claude --plugin-dir ./justtranscribe-mcp.
Tools
transcribe_url— start transcribing a public video/audio link (YouTube, TikTok, Instagram, Facebook, Pinterest, Google Drive). Returns an id; processing takes minutes.get_transcript— poll by id; returns the full text once complete (optionally timestamped segments and the AI analysis).list_transcripts— the account's transcripts, newest first.export_transcript— TXT, SRT, VTT, Markdown or CSV as text.delete_transcript— permanently remove a transcript and its media.get_account— identify the key's account and limits.
Limits
Public links only (no private or login-required videos, no live streams), up to 150 minutes per recording. The server enforces the same limits as the app and the REST API: 120 requests/minute per key, 3 new transcripts/hour, one processing at a time during the beta. File upload isn't possible over MCP — use the app or the REST API.
Environment
JUSTTRANSCRIBE_API_KEY(required) — yourjt_live_…key.JUSTTRANSCRIBE_API_URL(optional) — API origin override; defaults tohttps://justtranscribe.ai.
MIT © Bramonti Ventures · Docs: https://justtranscribe.ai/mcp
This server cannot be deployed
Maintenance
Related MCP Connectors
Get transcripts from YouTube, TikTok, X, Instagram and more - even when captions are off.
Multimodal video analysis MCP — transcription, vision, and OCR for any video URL.
Any video URL to LLM-ready transcript. ASR built in, no captions needed. TikTok, X, TED and more.
Any social-video URL → transcript, metadata, frames, OCR, summary, search, Q&A. MCP server + x402.
Related MCP Servers
- AlicenseAqualityAmaintenanceTranscribes videos from 1000+ platforms (YouTube, TikTok, Vimeo, etc.) and local video files using OpenAI's Whisper model, with support for 90+ languages and multiple output formats.814 npm6MIT
- FlicenseAqualityDmaintenanceEnables high-quality transcription and subtitle generation from local media files or URLs using Faster Whisper on local hardware. It supports automatic language detection and integration with MCP clients for seamless speech-to-text workflows.3-
- AlicenseNot gradedqualityBmaintenanceFetches YouTube video transcripts with timestamps and provides them to LLM agents via MCP, enabling natural language access to video content.58 npm4MIT
- FlicenseAqualityCmaintenanceProvides YouTube video transcripts (text and metadata) as MCP tools for AI hosts like Claude Code, enabling summarization, search, and Q&A without watching the video.2-