
wordtube
io.github.CloudCorpRecords/wordtube
Find YouTube videos for learning Japanese or English, by level and word.
Documentation
wordtube
Learn a language from real video, without leaving your MCP client. wordtube is a small local forwarding client for the hosted LingoDew service: it answers tools/list on your own machine from the captured tool definitions and forwards every lookup upstream, returning the results exactly as the service sends them back.
The surface is read-only catalogue data for Japanese and English learners: videos matched by level, topic and keywords, statistics on how a word is actually used across a video corpus, measured stats for YouTube channels in the library, and basic product facts. Nothing is cached or computed locally — the package is a pipe, not a catalogue.
The tools
find_videos— Find real YouTube videos for learning Japanese or English.languageis required ("japanese"or"english"). Optional filters:level(JLPT N5–N1 for Japanese, CEFR A1–C2 for English),topicsuch as Food or Travel,querymatched against video titles and channel names, andlimit(1–25, default 10). Results include titles, channels, levels and links to video pages with tappable subtitles.word_in_videos— How often a word is heard across the video corpus: number of videos, total occurrences, coverage, frequency rank and level, plus a word page with native-audio examples. Bothlanguageandwordare required; for Japanese the word may be a headword, kana reading or romaji.channel_stats— Measured stats for a YouTube channel in the library: its video count, and per level its speaking speed, rank among channels at that level, share of its videos and main topics. Bothlanguageandchannelare required; the channel may be given as a name, channel id or URL.about_lingodew— Product facts about LingoDew itself: features, platforms, pricing and the free tier. Takes no arguments.
Install and run
As an MCP server via npm:
npx -y @reneturcios/wordtube mcp
MCP client configuration:
{
"mcpServers": {
"wordtube": {
"command": "npx",
"args": ["-y", "@reneturcios/wordtube", "mcp"]
}
}
}
As a container:
docker run -i ghcr.io/cloudcorprecords/wordtube
docker run --rm ghcr.io/cloudcorprecords/wordtube find_videos '{"language":"japanese","level":"N5"}'
CLI
One binary serves both the command line and the MCP server. Each tool is a subcommand taking a single JSON string with that tool's arguments:
wordtube find_videos '{"language":"english","topic":"Food"}'
wordtube word_in_videos '{"language":"japanese","word":"食べる"}'
wordtube channel_stats '{"language":"english","channel":"Some Channel"}'
wordtube about_lingodew '{}'
wordtube --version
Exit codes: 0 when the forwarded call succeeds, 1 when the upstream result is an error, and 2 for usage problems such as an unknown subcommand or a malformed JSON argument.
Configuration
LINGODEW_MCP_URL— Overrides the upstream endpoint. The default is the pinned hosted endpoint,https://www.lingodew.com/mcp. (The override variable is a reconstruction added by this rebuild.)
The LingoDew registry record declares no headers on its remote, so wordtube sends no authentication header of any kind — lookups are keyless.
Upstream failures — a JSON-RPC error from the service, a non-2xx response, an unreadable body, or a request that fails outright — are reported to the caller as an error result whose text starts with wordtube error:.
How forwarding works
The first forwarded call opens a session: an initialize request (protocol version 2025-03-26), followed by a notifications/initialized notification, then the tools/call itself. If the service answers the initialize request with an Mcp-Session-Id header, that value is carried on every later request, and the open session is reused for the rest of the process. Replies are accepted either as plain JSON or as server-sent events. Successful results pass through verbatim, including any isError flag and any extra fields the service includes.
Development
npm install
npm run build # tsc -> dist/ (dist/cli.js is the bin)
npm test # tsx --test test/*.test.ts — stubbed upstream, no network
npm run dev # run the CLI from source
The test suite checks the served tool list against the captured surface, exercises the session handshake and its reuse, asserts that no Authorization header is sent under any environment, and verifies that arguments go upstream untouched and results come back untouched.
License
MIT — see LICENSE. Copyright (c) 2026 Rene Turcios.
@reneturcios/wordtubenpm install @reneturcios/wordtubeghcr.io/cloudcorprecords/wordtube:0.1.0docker pull ghcr.io/cloudcorprecords/wordtube:0.1.0:undefinedRelated Servers
io.github.CloudCorpRecords/abandoncheck
Judge whether a dependency is maintained, from factual signals: releases, commits, archival.
io.github.CloudCorpRecords/actionbridge
Zapier exposed actions as MCP tools via a generic action runner.
io.github.CloudCorpRecords/actionpin
GitHub Actions supply-chain hardener: audit workflows for unpinned actions and build a SHA pin plan.