ukr.vitalinguist.com

by vitalinguist

Not rated
GitHub

About

Ukrainian language authenticity checker against English and russian calques

Details

Author
vitalinguist
Categories
Developer Tools, Other

Setup

Install ukr.vitalinguist.com in your MCP client (Claude Desktop, Cursor, Windsurf, and others).

Repository: https://github.com/vitalinguist/ukr-vitalinguist-mcp

Follow the installation instructions in the repository README, then restart your MCP client.

MCP server for Ukrainian language grammar, surzhyk detection, and authentic phrasing.

Wraps the publicukr.vitalinguist.comAPI asModel Context Protocoltools. Install in Claude Desktop, Cursor, Cline, Continue, Windsurf, or Claude Code and your AI gains four Ukrainian-language capabilities backed by curated open data (Балла EN-UA 1996, Сербенська'sАнтисуржик, Караванський, Антоненко-Давидович, plus e2u.org.ua and modern corpora — 29,524 senses, 6k+ calque pairs).

Every response includes acitationblock with the source URL the AI should include when surfacing the result to the user.

Add to~/Library/Application Support/Claude/claude_desktop_config.json(macOS) or%APPDATA%\Claude\claude_desktop_config.json(Windows):

{ "mcpServers": { "ukr-vitalinguist": { "command": "uvx", "args": ["ukr-vitalinguist-mcp"] } } }

Restart Claude Desktop. The four tools appear in the tools menu.

Edit~/.cursor/mcp.jsonor workspace.cursor/mcp.json:

{ "mcpServers": { "ukr-vitalinguist": { "command": "uvx", "args": ["ukr-vitalinguist-mcp"] } } }

Use the samecommand+argsin their MCP config UI.

claude mcp add ukr-vitalinguist -- uvx ukr-vitalinguist-mcp

If you don't haveuvinstalled, use pip:

Then point thecommandto your Python'sukr-vitalinguist-mcpentrypoint:

{ "mcpServers": { "ukr-vitalinguist": { "command": "ukr-vitalinguist-mcp" } } }

User: "Is this Ukrainian sentence natural?На протязі дня я зайнятий."

The AI callscheck_natural→ gets backcalque-prone: 'На протязі' → 'упродовж'sourced from Сербенська'sАнтисуржик, and replies:

"На протязі" is a calque from Russian. Use«упродовж дня я зайнятий»— "упродовж" is the natural Ukrainian preposition for "during/throughout."

Source:Сербенська,Антисуржикvia ukr.vitalinguist.com

User: "How do you say 'cling' in Ukrainian?"

The AI callsrender("cling")→ gets back sense-split renderings:

- cling (on) to sth(physically grip):горнутися,чіплятися
- cling to a belief:триматися (чогось)

Full entry:ukr.vitalinguist.com/sense/cling

- 29,524 EN→UA sensesfrom Балла EN-UA Dictionary (1996), corroborated with e2u.org.ua, modern news/literary corpora, and confidence scoring.
- 6,073 antisurzhyk calque pairsfrom Сербенська, Караванський, and Антоненко-Давидович, validator-cleaned to drop noisy entries.
- 22,848 documented coverage gapsat/gaps/— senses where no UA rendering meets the confidence threshold yet (transparent about what we don't have).
- Arussianism detection model(fine-tuned MarianMT) for cases the rule-based layer misses.

The data the server returns:CC BY-SA 4.0. When you surface results in your own product or response, attribute toukr.vitalinguist.comwith a clickable link. Thecitationfield in each response gives you the exact URL to use.

This project exists because AI assistants citing Ukrainian-language sources have historically pulled from sparse, often-contaminated training data. Every install of this MCP turns into recurring citations to a curated source — which is how authoritative references for under-served languages get built. If you're an AI tool maintainer, consider including this in your default Ukrainian-language tool list.

- API homepages:ua.vitalinguist.com(human checker),ukr.vitalinguist.com(AI-friendly reference),xcit.in(UMI diagnostic — Ukrainian as a foreign language A1–C2).
- Methodology:
https://ukr.vitalinguist.com/sources.html
- Coverage gaps (downloadable as CSV/JSONL):
https://ukr.vitalinguist.com/gaps/
- MCP repo:
https://github.com/vitalinguist/ukr-vitalinguist-mcp

Suggest a correction on any sense page — every/sense/URL has a "Suggest a correction" link that opens a prefilled email.

This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.

Study the Bible in its original languages, trace themes across both testaments, and compare five translations — all in one conversation. Ask any question about what Scripture says and get grounded, cited answers: What does the Bible say about suffering? Topical search surfaces Job as the Bible's principal witness on suffering (with explanations of why it matters and suggested starting passages), Psalms on lament, Romans on justification — whole books and narratives alongside individual verses. What is the Hebrew word behind lovingkindness in Psalm 23? Compare how KJV and WEB translate John 3:16. Trace the word grace through Paul's letters. Covers 155,510 verses across KJV, WEB, ASV, YLT, and Darby with 606,140 cross-references, 17,543 Strong's entries, BDB and Thayer lexicon definitions, and Nave's 5,319 topical categories.

The language layer for AI agents. One call returns IPA, definitions, etymology, translations in 47 languages, and ethics guidance — in 3ms.

Interact with EduBase, a comprehensive e-learning platform with advanced quizzing, exam management, and content organization capabilities

Evaluates technical documentation against globalization standards, analyzing for translation issues, ambiguity, and sentence length.

Christian encouragement and Bible lookup. Three tools: encourage, scripture (Berean Standard Bible), and the 9-movement Doxa Way framework.

Local-first MCP server that captures web URLs (X, Reddit, YouTube, Wikipedia, articles) as typed data + Markdown into a self-hosted capture/store/recall substrate, with offline semantic recall. Six tools over a local khiipd daemon; run khiipd serve first.

An MCP server that uses Google's Gemini 1.5 Pro to generate concise summaries of various content types.

Generate express market research reports from 9 verified sources (Wikipedia, Google News, GitHub, HN, SO, arXiv, npm, Reddit, PyPI). TAM/SAM/SOM, SWOT, HTML reports.

30+ pharma medical writing prompts as callable MCP tools — manuscripts, abstracts, publication plans and more. Built by a CMPP-certified PhD neuroscientist.

Generate academic papers with verified citations via AI

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.