Transcribevideototext Sdk
LocalAbout Transcribevideototext Sdk
Transcribevideototext Sdk provides a typed JavaScript/TypeScript client, CLI, and MCP server for the VideoToText transcription API. The packages share API-key authentication, and the MCP server can transcribe public links from supported sites as well as media inputs.
Key features
- 01Typed JavaScript/TypeScript client
- 02Transcription CLI
- 03MCP server for transcription
- 04Shared vtt_ API key authentication
- 05Transcription from public media links
Use cases
- 01Transcribe a local interview video with diarization
- 02Transcribe a podcast from its URL using the client
- 03Transcribe public video links through an MCP client
MCP support
The MCP server connects MCP clients to the VideoToText transcription API. It supports transcription from media and public links, including YouTube, X, and LinkedIn.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEtranscribevideototext-sdk
Official client, CLI, and MCP server for the VideoToText
transcription API. All three are thin clients of the public REST API and authenticate with the
same vtt_ API key.
| Package | Install | What it is |
|---|---|---|
@transcribevideototext/client | npm i @transcribevideototext/client | Typed JS/TS client. |
@transcribevideototext/cli | npm i -g @transcribevideototext/cli | The transcribe command. |
@transcribevideototext/mcp-server | npx @transcribevideototext/mcp-server | MCP server for Claude Code, Cursor, Codex. |
Get an API key from the dashboard → Developers → API Keys.
Quick start
# CLI
npm install -g @transcribevideototext/cli
transcribe login
transcribe ./interview.mp4 --diarize
# MCP (Claude Code)
claude mcp add vtt -e VTT_API_KEY=vtt_… -- npx -y @transcribevideototext/mcp-server
Through the MCP server you can also paste a YouTube, X, LinkedIn (and many other popular sites) link (public content): it's downloaded locally — the downloader is auto-fetched on first use — then transcribed. See the MCP server README.
// Client
import { createClient } from "@transcribevideototext/client";
const client = createClient({ apiKey: process.env.VTT_API_KEY! });
const result = await client.transcribeAndWait({ url: "https://example.com/podcast.mp3" });
console.log(result.segments.map((s) => s.text).join(" "));
Development
pnpm install
pnpm build # builds all three (client first)
pnpm typecheck
pnpm lint
Release
The three packages version in lockstep via Changesets.
pnpm changeset # describe a change
pnpm version-packages # bump versions
pnpm release # build + publish to npm
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.