@pipeworx/wiktionary

Wiktionary MCP — multilingual dictionary content via the MediaWiki REST API. ~6M entries across hundreds of languages. No auth.

Part of Pipeworx — an MCP gateway connecting AI agents to 1394+ live data sources.

Tools

• definition(word, lang?) — parsed definitions grouped by language section (uses Wikimedia's rest_v1/page/definition)

• summary(word, lang?) — page summary

• search(query, lang?, limit?) — title-prefix + fulltext search

• etymology(word, lang?) — extracts the Etymology section

• pronunciations(word, lang?) — extracts IPA pronunciation rows

Related MCP server: word-orb

Languages

lang is the Wiktionary subdomain code: en (default), fr, de, ja, zh, es, ...

Data source

• https://<lang>.wiktionary.org/api/rest_v1/

• https://<lang>.wiktionary.org/w/api.php (for search + wikitext extraction)

Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

{
  "mcpServers": {
    "wiktionary": {
      "url": "https://gateway.pipeworx.io/wiktionary/mcp"
    }
  }
}

Or connect to the full Pipeworx gateway for access to all 1394+ data sources:

{
  "mcpServers": {
    "pipeworx": {
      "url": "https://gateway.pipeworx.io/mcp"
    }
  }
}

Using with ask_pipeworx

Instead of calling tools directly, you can ask questions in plain English:

ask_pipeworx({ question: "your question about Wiktionary data" })

The gateway picks the right tool and fills the arguments automatically.

More

• Docs and guides

• pipeworx.io

License

MIT