BESSER-MCP-Server
Click on "Install 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., "@BESSER-MCP-ServerCreate a model for a task management app with User and Task classes"
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.
BESSER-MCP-Server
A minimal Model Context Protocol (MCP) server implementation for BESSER, the Python-based low-code modeling platform.
Overview
This MCP server provides access to BESSER's low-code platform modeling and code generation capabilities through the Model Context Protocol, enabling AI assistants and other MCP clients to interact with BESSER's features.
Related MCP server: MCP Tool Factory
Features
The BESSER MCP Server can be used both locally or deployed on a server.
To deploy it, the --dist or -d option needs to be added when starting the server
Both the local and distant implementation open the server on the following address: http://127.0.0.1:8000/sse.
However, the port can be changed through the --port or -p option.
Help on the server option can be printed using the --help or -h option.
The BESSER MCP Server provides the following tools:
Information tools:
about: Get information about BESSER and this MCP serverget_model_info: Get detailed information about an existing domain model (classes, attributes, methods, relationships)
Creation tools:
new_model: Create a new B-UML DomainModel with a specified nameadd_class: Add a new class in the modeladd_method_to_class: Add a method to a given classadd_attribute_to_class: Add an attribute to a given classadd_binary_association: Add a binary association between two classesadd_association_class: Add an association class to an existing associationadd_enumeration: Add an enumeration to the modeladd_enumeration_literal: Add a literal to an enumerationadd_generalization: Add a generalization relation between two classesadd_ocl_constraint: Add an OCL constraint to a class
Deletion tools:
delete_class: Remove a class from the modeldelete_method_to_class: Remove a method from a given classdelete_attribute_to_class: Remove an attribute from a given classdelete_binary_association: Remove a binary association between two classesdelete_association_class: Remove an association class from an existing associationdelete_enumeration: Remove an enumeration from the modeldelete_enumeration_literal: Remove a literal from an enumerationdelete_generalization: Remove a generalization relation between two classesdelete_ocl_constraint: Remove an OCL constraint
Generation tools:
sql_generation: Generate the SQL representation of the modelpython_generation: Generate the set of Python classes implementing the modelbackend_generation: Generate a backend implementation for the modeljava_generation: Generate the set of Java classes implementing the modeljson_schema_generation: Generate the JSON Schema of the modeljson_smart_data_generation: Generate the Smart Data Schema for the modelpydantic_classes_generation: Generate a Pydentic representation of the modelrdf_generation: Generate the RDF vocabulary for the modelrest_api_generation: Generate a Rest API for the modelsql_alchemy_generation: Generate the SQLAlchemy code for the model
If the server is started locally, the server will directly access the file system for file generation. If the server is distant, the generated code will be sent back to the LLM. For this reason, generators producing multiple files are currently disabled in the distant mode.
Prerequisites
Before using the MCP server, ensure you have:
Python 3.8+ installed
Required packages:
pip install mcp besserOr if using UV:
uv add mcp besser
Usage
Running the Server
Production Mode
Run the MCP server directly:
python src/besser_mcp_server/server.pyOr install the package and run it:
pip install -e .
python -m besser_mcp_server.serverDevelopment Mode
For development and debugging, use the MCP development server which includes the MCP inspector:
mcp dev src/besser_mcp_server/server.pyThis will:
Start the server in development mode
Launch the MCP inspector in your browser for interactive testing
Provide detailed logging and debugging information
Allow you to test tools directly without configuring a client
Configuring MCP Clients
To use this server with MCP clients like Claude Desktop, Cursor and Cline, you need to configure them to connect to the server.
Recommended setup (Cline)
Open the Cline panel on VScode
Click on "Manage MCP Server" at the bottom, then the cog
Clicking on "Configure MCP server" will open the configuration file
Add the "besser-mcp-server" in the "mcpServers" section:
{ "mcpServers": { "besser-mcp": { "url": "http://127.0.0.1:8000/sse" } } }
Quick Setup (Windows)
Use the provided PowerShell installation script:
# For Cursor IDE
.\install-mcp-config.ps1 -Client cursor
# For Claude Desktop
.\install-mcp-config.ps1 -Client claude
# For both clients
.\install-mcp-config.ps1 -Client bothManual Configuration
Cline
Open the Cline panel on VScode
Click on "Manage MCP Server" at the bottom, then the cog
Clicking on "Configure MCP server" will open the configuration file
Add the "besser-mcp-server" in the "mcpServers" section:
{ "mcpServers": { "besser-mcp": { "url": "http://127.0.0.1:8000/sse" } } }
Claude Desktop
Locate the configuration directory:
macOS:
~/Library/Application Support/Claude/Windows:
%APPDATA%\Claude\
Create or edit
claude_desktop_config.json:{ "mcpServers": { "besser-mcp": { "type": "sse", "url": "http://127.0.0.1:8000/sse" } } }
Cursor IDE
Create the Cursor MCP configuration directory:
mkdir -p ~/.cursor # macOS/Linux # or mkdir %USERPROFILE%\.cursor # WindowsCreate or edit
~/.cursor/mcp.json:{ "mcpServers": { "besser-mcp": { "url": "http://127.0.0.1:8000/sse" } } }
Distant mode for collaboration
By using and sharing a URL where the model can be saved multiple users can edit the same model.
To do so, you have to deploy the MCP server on distant mode using the -d or --dist option.
You can change the port used for this server using the -p or --port option followed by the port number.
With this server, users can change the model by giving the URL as parameter. Example prompt:
Add a class representing users in the model available at the URL https://my-domain.com/modelDisclaimer: the distant mode currently has limitations regarding concurrent access and multi-file generation
Testing the Connection
After configuring your MCP client:
Start the server
Restart your MCP client
Test the connection by asking:
"What is BESSER?" (uses the
abouttool)"Create a new domain model called 'MyModel'" (uses
new_model)"Add a class named 'Person' to the model" (uses
add_class)"Show me information about the domain model" (uses
get_model_info)"Generate SQL from the domain model" (uses
sql_generation)
Configuration Files
The repository includes several pre-configured files to help with setup:
mcp-config.json- Basic cross-platform configuration templatemcp-config-examples.json- Multiple configuration examplesclaude-desktop-config-windows.json- Windows-specific Claude Desktop configcursor-mcp-config-windows.json- Windows-specific Cursor configinstall-mcp-config.ps1- Automatic installation script for Windows
Documentation
For detailed setup instructions and troubleshooting:
QUICK-START.md- Fast setup guideMCP-CLIENT-SETUP.md- Comprehensive setup instructions with troubleshooting
Example Usage
Once connected, you can interact with BESSER through natural language:
User: "Create a new domain model called 'ECommerceSystem'"
Assistant: [Uses new_model tool to create the model]
User: "Add a Customer class to the ECommerceSystem model"
Assistant: [Uses add_class tool to add the Customer class]
User: "Add a Product class as well"
Assistant: [Uses add_class tool to add the Product class]
User: "Show me information about the current domain model"
Assistant: [Uses get_model_info tool to display model structure, classes, and relationships]
User: "Generate SQL from this domain model"
Assistant: [Uses sql_generation tool to create SQL representation]
User: "What is BESSER?"
Assistant: [Uses about tool to provide information about BESSER]Troubleshooting
Common Issues
"mcp is not found" or import errors:
Make sure you have the correct MCP package installed:
pip install mcpIf you have an older version, upgrade it:
pip install --upgrade mcpVerify installation:
python -c "import mcp.server; print('MCP installed correctly')"
"Command not found" errors: Ensure Python is in your system PATH
"Permission denied" errors: Ensure the server script has appropriate permissions
License
This project is licensed under the MIT License - see the LICENSE file for details.
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/BESSER-PEARL/BESSER-MCP-Server'
If you have feedback or need assistance with the MCP directory API, please join our Discord server