๐ mhlabs-mcp-tools
mhlabs-mcp-tools is a Modular MCP Tools Server built using FastMCP. It offers an extendable AI tool ecosystem categorized by functions (Text Preprocessing, NLP Components, Document Analysis, etc.). These tools can be dynamically loaded and served via MCP (Model Context Protocol) through STDIO transport.
This project is part of the MHLabs AI Agentic Ecosystem and is designed to work with mhlabs-mcp-server, mhlabs-mcp-agents, and downstream A2A agent frameworks.
๐ Quick Start
Development Setup
- Clone and Navigate:
cd src/mhlabs_mcp_tools - Install Dependencies:
pip install -r requirements.txt - Configure Environment:
cp .env.example .env # Edit .env with your configuration - Start the Server:
# Default STDIO transport (for local MCP clients) python mcp_server.py # HTTP transport (for web-based clients) python mcp_server.py --transport http --port 9000 or after installed mhlabs-mcp-tools python -m mhlabs_mcp_tools.mcp_server --transport http --port 9000 # Using FastMCP CLI (recommended) fastmcp run mcp_server.py -t streamable-http --port 9000 -l DEBUG # Debug mode with authentication disabled python mcp_server.py --transport http --debug --no-auth
Transport Options
1. STDIO Transport (default)
- ๐ง Ideal for: Local tools, command-line integrations, Claude Desktop
- ๐ Usage:
python mcp_server.pyorpython mcp_server.py --transport stdio
2. HTTP (Streamable) Transport
- ๐ Ideal for: Web-based deployments, microservices, remote access
- ๐ Usage:
python mcp_server.py --transport http --port 9000 - ๐ URL:
http://127.0.0.1:9000/mcp/
3. SSE Transport (deprecated)
- โ ๏ธ Only for legacy support - use HTTP transport for new projects
- ๐ Usage:
python mcp_server.py --transport sse --port 9000
FastMCP CLI Usage
# Standard HTTP server
fastmcp run mcp_server.py -t streamable-http --port 9000 -l DEBUG
# With custom host
fastmcp run mcp_server.py -t streamable-http --host 0.0.0.0 --port 9000 -l DEBUG
# STDIO transport (for local clients)
fastmcp run mcp_server.py -t stdio
# Development mode with MCP Inspector
fastmcp dev mcp_server.py -t streamable-http --port 9000
VS Code Development
- Open in VS Code:
code . - Use Debug Configurations:
Debug MCP Server (STDIO): Run with STDIO transportDebug MCP Server (HTTP): Run with HTTP transportDebug Tests: Run the test suite
โจ Features
- FastMCP Server: A pure FastMCP implementation that supports multiple transport protocols.
- Factory Pattern: A reusable MCP tools factory for easy service management.
- Domain-Based Organization: Services are organized by business domains (HR, Tech Support, etc.).
- Authentication: Optional Azure AD authentication support.
- Multiple Transports: Supports STDIO, HTTP (Streamable), and SSE transports.
- VS Code Integration: Comes with debug configurations and development settings.
- Comprehensive Testing: Unit tests using pytest.
- Flexible Configuration: Environment-based configuration management.
๐ง Technical Details
Architecture
mhlabs_mcp_tools/
โโโ .gitignore
โโโ .vscode/
โ โโโ settings.json
โโโ CHANGELOG.md
โโโ LICENSE
โโโ README.md
โโโ docs/
โ โโโ index.md
โโโ examples/
โ โโโ example_client.py
โ โโโ example_client_http.py
โโโ mkdocs.yml
โโโ pyproject.toml
โโโ requirements.txt
โโโ server.json
โโโ src/
โโโ __init__.py
โโโ main.py
โโโ mhlabs_mcp_tools/
โโโ __init__.py
โโโ core/
โ โโโ __init__.py
โ โโโ config.py
โ โโโ constants.py
โ โโโ factory.py
โ โโโ prompts.py
โโโ data/
โ โโโ __init__.py
โ โโโ external/
โ โ โโโ __init__.py
โ โโโ interim/
โ โ โโโ __init__.py
โ โโโ processed/
โ โ โโโ __init__.py
โ โโโ raw/
โ โโโ __init__.py
โ โโโ contractions_dict.json
โ โโโ custom_substitutions.csv
โ โโโ leftovers_dict.json
โ โโโ slang_dict.json
โโโ handlers/
โ โโโ __init__.py
โ โโโ custom_exceptions.py
โ โโโ output_generator.py
โโโ mcp_server.py
โโโ models/
โ โโโ __init__.py
โโโ nlp_components/
โ โโโ __init__.py
โ โโโ nlp_model.py
โโโ services/
โ โโโ __init__.py
โ โโโ langchain_framework.py
โ โโโ spacy_extractor.py
โโโ text_preprocessing/
โโโ __init__.py
โโโ contractions.py
โโโ emo_unicode.py
โโโ slang_text.py
โโโ text_preprocessing.py
๐ป Usage Examples
Available Services
Currently, the package is organized into three primary modules:
1. NLP Components
| Component Type | Description |
|---|---|
| tokenize | Text tokenization |
| pos | Part-of-Speech tagging |
| lemma | Word lemmatization |
| morphology | Study of word forms |
| dep | Dependency parsing |
| ner | Named Entity Recognition |
| norm | Text normalization |
2. Text Preprocessing
This module provides users with a wide range of text preprocessing tools:
| Function | Description |
|---|---|
| to_lower | Convert text to lowercase |
| to_upper | Convert text to uppercase |
| remove_number | Remove numerical characters |
| remove_itemized_bullet_and_numbering | Eliminate itemized/bullet-point numbering |
| remove_url | Remove URLs from text |
| remove_punctuation | Remove punctuation marks |
| remove_special_character | Remove special characters |
| keep_alpha_numeric | Keep only alphanumeric characters |
| remove_whitespace | Remove excess whitespace |
| normalize_unicode | Normalize Unicode characters |
| remove_stopword | Eliminate common stopwords |
| remove_freqwords | Remove frequently occurring words |
| remove_rarewords | Remove rare words |
| remove_email | Remove email addresses |
| remove_phone_number | Remove phone numbers |
| remove_ssn | Remove Social Security Numbers (SSN) |
| remove_credit_card_number | Remove credit card numbers |
| remove_emoji | Remove emojis |
| remove_emoticons | Remove emoticons |
| convert_emoticons_to_words | Convert emoticons to words |
| convert_emojis_to_words | Convert emojis to words |
| remove_html | Remove HTML tags |
| chat_words_conversion | Convert chat language to standard English |
| expand_contraction | Expand contractions (e.g., "can't" to "cannot") |
| tokenize_word | Tokenize words |
| tokenize_sentence | Tokenize sentences |
| stem_word | Stem words |
| lemmatize_word | Lemmatize words |
| preprocess_text | Combine multiple preprocessing steps into one function |
MCP Client Usage
Python Client
import asyncio
from fastmcp import Client
client = Client("http://localhost:9000/mcp")
async def main():
async with client:
tools = await client.list_tools()
# tools -> list[mcp.types.Tool]
# print(tools)
for tool in tools:
print(f"Tool: {tool.name}")
result = await client.call_tool("textprep.expand_contraction", {"input_text": "The must've SSN is 859-98-0987. The employee's phone number is 555-555-5555."})
print("Result:", result)
asyncio.run(main())
Command Line Testing
# Test the server is running
curl http://localhost:9000/mcp/
# With FastMCP CLI for testing
fastmcp dev mcp_server.py -t streamable-http --port 9000
Quick Test
Test STDIO Transport:
# Start server in STDIO mode
python mcp_server.py --debug --no-auth
# Test with client_example.py
python client_example.py
Test HTTP Transport:
# Start HTTP server
python mcp_server.py --transport http --port 9000 --debug --no-auth
# Test with FastMCP client
python -c "
from fastmcp import Client
import asyncio
async def test():
async with Client('http://localhost:9000/mcp') as client:
result = await client.call_tool("textprep.expand_contraction", {"input_text": "The must've SSN is 859-98-0987. The employee's phone number is 555-555-5555."})
print(result)
asyncio.run(test())
"
๐ Documentation
Configuration
Environment Variables
Create a .env file based on .env.example:
# Server Settings
MCP_HOST=0.0.0.0
MCP_PORT=9000
MCP_DEBUG=false
MCP_SERVER_NAME=MHLABS MCP Server
# Authentication Settings
MCP_ENABLE_AUTH=true
AZURE_TENANT_ID=your-tenant-id-here
AZURE_CLIENT_ID=your-client-id-here
AZURE_JWKS_URI=https://login.microsoftonline.com/your-tenant-id/discovery/v2.0/keys
AZURE_ISSUER=https://sts.windows.net/your-tenant-id/
AZURE_AUDIENCE=api://your-client-id
Authentication
When MCP_ENABLE_AUTH=true, the server expects Azure AD Bearer tokens. Configure your Azure App Registration with the appropriate settings.
For development, set MCP_ENABLE_AUTH=false to disable authentication.
Adding New Services
- Create Service Class:
from core.factory import MCPToolBase, Domain class MyService(MCPToolBase): def __init__(self): super().__init__(Domain.MY_DOMAIN) def register_tools(self, mcp): @mcp.tool(tags={self.domain.value}) async def my_tool(param: str) -> str: # Tool implementation pass @property def tool_count(self) -> int: return 1 # Number of tools - Register in Server:
# In mcp_server.py (gets registered automatically from services/ directory) factory.register_service(MyService()) - Add Domain (if new):
# In core/factory.py class Domain(Enum): # ... existing domains MY_DOMAIN = "my_domain"
Server Arguments
usage: mcp_server.py [-h] [--transport {stdio,http,streamable-http,sse}]
[--host HOST] [--port PORT] [--debug] [--no-auth]
MHLABS MCP Server
options:
-h, --help show this help message and exit
--transport, -t Transport protocol (default: stdio)
--host HOST Host to bind to for HTTP transport (default: 127.0.0.1)
--port, -p PORT Port to bind to for HTTP transport (default: 9000)
--debug Enable debug mode
--no-auth Disable authentication
๐ License
MIT License ยฉ 2025 MusaddiqueHussain Labs
๐ค Contributing
- Follow the existing code structure and patterns.
- Add tests for new functionality.
- Update documentation for new features.
- Use the provided VS Code configurations for development.
๐ง Learn More
- MCP Protocol: https://modelcontextprotocol.io
- FastMCP GitHub: https://github.com/fastmcp/fastmcp
- LangGraph Integration Guide (coming soon)
๐ก Usage Tip
If you want to embed mhlabs-mcp-tools into a larger MCP-based orchestrator:
from fastmcp import StdioServerParameters
server_params = StdioServerParameters(
command="python",
args=["-m", "mhlabs_mcp_tools.server"],
//env={"MHLABS_MCP_CATEGORY": "textprep,nlp"}
)
Developed with โค๏ธ by MusaddiqueHussain Labs








