MCP Arabic Toolkit
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MCP Arabic Toolkit normalize this Arabic text and strip diacritics: الْعَرَبِيَّةُ"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP Arabic Toolkit 🌍
A high-performance Model Context Protocol (MCP) server providing Arabic Natural Language Processing (NLP), text normalization, diacritics stripping, dialect detection, and token efficiency analysis for AI agents.
Developed by Tarek Mohamed (@seotarek)
✨ Features
⚡ Model Context Protocol (MCP) Compatible: Plug directly into Claude Desktop, Cursor, and any MCP client.
🧹 Advanced Arabic Normalization: Unify complex Alef variants, Hamza, Yaa, Taa Marbuta, and strip Tatweel.
🎯 Tashkeel Stripping: Remove diacritics (Harakat) accurately with zero distortion to core roots.
🗣️ Arabic Dialect Classification: Heuristic and marker-based identification across Egyptian (EGY), Levantine (LEV), Gulf (GLF), Maghrebi (MAG), and Modern Standard Arabic (MSA).
📊 Token & Prompt Efficiency Metrics: Calculate character-to-word and token compression statistics to optimize LLM prompt budgets for Arabic workloads.
Related MCP server: flexorch-mcp
🛠️ MCP Tools Exposed
Tool Name | Description |
| Normalizes Arabic letters, unifies orthographic variants, and strips elongation. |
| Strips all Tashkeel / Harakat from the input text. |
| Identifies regional dialect with confidence scores. |
| Measures Arabic prompt density, word counts, and estimated token ratios. |
🚀 Quickstart & Installation
Option A: Direct Run via uvx (No installation needed)
uvx mcp-arabic-toolkitOr configure Claude Desktop / Cursor directly:
{
"mcpServers": {
"arabic-toolkit": {
"command": "uvx",
"args": ["mcp-arabic-toolkit"]
}
}
}Option B: Local Clone & Development
git clone https://github.com/seotarek/mcp-arabic-toolkit.git
cd mcp-arabic-toolkit
pip install -e .🧪 Testing
Run the automated test suite:
python -m unittest discover tests📜 License
This project is open source and available under the MIT License.
Available Tools
4 toolsarabic_detect_dialectA
Classifies Arabic text into Modern Standard Arabic (MSA), Egyptian, Levantine, Gulf, or Maghrebi.
Inputs: text (required).
arabic_normalize_textB
Normalizes Arabic text by unifying character variants (Alef, Yaa), removing Tatweel, and stripping diacritics.
Inputs: text (required), unify_taa, unify_yaa, unify_alef.
arabic_strip_diacriticsB
Removes all vowel marks (Tashkeel/Harakat) and elongation (Tatweel) from Arabic text.
Inputs: text (required).
arabic_token_metricsB
Computes token-to-word ratio, character count, and compression statistics for Arabic prompts.
Inputs: text (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 4 tools
arabic_normalize_text and arabic_strip_diacritics overlap substantially, since normalization explicitly includes stripping diacritics and Tatweel, making strip_diacritics a near-subset. Descriptions do clarify that normalize additionally unifies Alef/Yaa variants, but an agent wanting only diacritic removal has two plausible choices. The other two tools (dialect detection, token metrics) are clearly distinct.
All four names use a consistent snake_case arabic_ prefix with a verb_noun structure (normalize_text, strip_diacritics, detect_dialect). arabic_token_metrics deviates slightly toward a noun phrase rather than an action verb, but the set reads predictably overall.
Four tools is a tight, well-scoped surface for a niche text-processing toolkit, with no filler operations. It is on the thin side — one or two more capabilities would round out the offering — but each existing tool earns its place.
The surface covers normalization, diacritic removal, dialect detection, and token statistics, but omits other core Arabic text operations such as tokenization/segmentation, stemming or stopword removal, and transliteration. Notably there is a metrics tool but no actual tokenizer, leaving a gap for prompt-processing workflows.
Maintenance
Related MCP Connectors
AI Visibility and Content Intelligence tools for Claude and MCP-compatible agents.
An MCP for marketers that gives agents tools for SEO, AEO, socials and ads data. Generous free tier; works in Claude, ChatGPT, Cursor and other MCP clients.
One MCP endpoint for Claude, GPT & Gemini: 100+ tools + no-code connectors + agent workers.
214 tools for AI agents over one MCP endpoint: image, video, audio, 3D, data, storage, mail, GPUs.
Related MCP Servers
- AlicenseAqualityDmaintenanceMCP server exposing Arabic text utilities: normalisation, tashkeel stripping, transliteration, heuristic dialect detection, and token counting.5MIT

flexorch-mcpofficial
AlicenseAqualityBmaintenanceEnables Claude and other MCP-compatible agents to process documents, extract structured data, detect PII, and export LLM-ready datasets through natural language tool calls.871 PyPI1MIT- AlicenseNot gradedqualityCmaintenanceProvides Arabic UI/UX design tools as an MCP server, offering RTL conversion, Arabic typography validation, cultural palettes, and anti-AI-slop auditing for AI-assisted design workflows.MIT
- AlicenseBqualityCmaintenanceEnables dependable Arabic text normalization and search preparation for RAG and deduplication, with typed outputs, stable fingerprints, and experimental dialect, sentiment, and diacritization tools.2MIT