Mcp Forge
MCP-Forge is a powerful framework for dynamically generating, managing, and monitoring Model Context Protocol (MCP) servers. Using the official MCP SDK, this tool enables you to create specialized MCP servers on demand through a centralized interface.
iddv
README
m# MCP-Forge: Dynamic MCP Server Generator and Manager
MCP-Forge is a powerful framework for dynamically generating, managing, and monitoring Model Context Protocol (MCP) servers. Using the official MCP SDK, this tool enables you to create specialized MCP servers on demand through a centralized interface.
Key Features
- Dynamic Server Generation: Create customized MCP servers from templates with specific capabilities
- Flexible Templating System: Extend base templates with custom handlers and server options
- Server Lifecycle Management: Start, stop, and monitor multiple MCP servers from a single control point
- Built-in Customizations: Add authentication, persistence, HTTP requests, database access, and more
- Advanced Monitoring: Track server resources, logs, and performance metrics
- Auto-Scaling: Automatically scale servers based on demand and resource usage
- Centralized Logging: Aggregate logs from all servers with filtering and searching
- Real-time Status Reporting: Monitor server health and performance in real-time
- Alerting System: Receive notifications for critical issues through various channels
- Comprehensive API: Full-featured client API for managing the entire server ecosystem
- MCP SDK Integration: Built on top of the official Model Context Protocol SDK
- Enterprise-Grade Security: Comprehensive protection mechanisms against common vulnerabilities and attacks
Installation
-
Clone the repository:
git clone https://github.com/iddv/mcp-forge.git cd mcp-forge -
Create a virtual environment:
python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate -
Install the required dependencies:
pip install -r requirements.txt
Quick Start
Start the Forge Server
python forge_mcp_server.py --port 9000
Create a Server
Use the client to create a new server:
python client.py create name="my-server" description="My custom MCP server" capabilities=["echo","time","uptime"]
List Available Servers
python client.py list --details
Start/Stop/Restart Servers
python client.py start <server-id>
python client.py stop <server-id>
python client.py restart <server-id>
Get Server Information
python client.py info <server-id>
python client.py logs <server-id>
python client.py stats <server-id>
Connect to a Specific Server
python client.py --port 9001 tools
Delete a Server
python client.py delete <server-id>
Customization
MCP-Forge supports extensive customization of generated servers:
Add Custom Handlers
python client.py create name="advanced-server" handlers=["file_reader","http_request"]
Configure Server Options
python client.py create name="storage-server" options={"persistence":"sqlite","auth":"basic"}
Advanced Features
Auto-Scaling
Configure auto-scaling for groups of servers:
python client.py call manage_auto_scaling action="create_group" group_name="web-servers" min_instances=1 max_instances=5
Resource Limits
Set resource limits for servers:
python client.py call set_resource_limit server_id="server-id" limit_name="cpu_percent" limit_value=50.0
Configuration Management
Get and set configuration options:
python client.py call get_config
python client.py call set_config section="server" key="log_level" value="debug"
Centralized Logging
Access and search logs from all servers:
python client.py call get_logs source="system" log_level="error" limit=50
python client.py call get_logs source="server-id" limit=100
Status Reporting
Get real-time status information for all servers:
python client.py call get_server_status
python client.py call get_server_status server_id="server-id"
Performance Metrics
Access performance metrics for the system or specific servers:
python client.py call get_metrics source="system" time_period="hour"
python client.py call get_metrics source="server-id" time_period="day"
Alert Management
Manage and respond to system alerts:
python client.py call get_alerts active_only=true
python client.py call acknowledge_alert alert_id="alert-id" user="admin"
python client.py call resolve_alert alert_id="alert-id" resolution_message="Fixed the issue"
Security Protection
MCP-Forge includes comprehensive security protection mechanisms:
# Generate a CSRF token for secure operations
python client.py call get_csrf_token session_id="your-session-id"
# Test the security protection mechanisms
python test_protection.py --host localhost --port 9000 --test all
Security Features
MCP-Forge implements a multi-layered security approach:
- Input Validation: Strict validation for all user inputs
- XSS Protection: Content Security Policy and input sanitization
- CSRF Protection: Token-based protection for state-changing operations
- SQL Injection Protection: Input validation against SQL patterns
- Rate Limiting: IP-based and global rate limiting
- DDoS Protection: Burst detection and IP blacklisting
- Security Headers: Content-Security-Policy, X-Frame-Options, etc.
- Server Hardening: Directory listing prevention, content length limits
- Data Encryption: Protection for sensitive data
- Intrusion Detection: Pattern-based detection of suspicious activities
- Audit Logging: Comprehensive security event logging
- Authentication: Role-based access control
For more details, see the Security Protection documentation.
Project Structure
mcp-forge/
├── forge_mcp_server.py # Core forge server
├── server_manager.py # Server instance management
├── config_manager.py # Configuration management
├── auto_scaler.py # Auto-scaling system
├── resource_monitor.py # Resource monitoring
├── process_monitor.py # Process monitoring
├── logging_system.py # Centralized logging system
├── log_aggregator.py # Log aggregation service
├── status_reporter.py # Status reporting system
├── metrics_collector.py # Performance metrics collection
├── alerting_system.py # Alerting system for critical issues
├── audit_logger.py # Security audit logging system
├── authentication_system.py # Authentication and authorization
├── protection_mechanisms.py # Security protection mechanisms
├── request_validator.py # Request validation and sanitization
├── quota_manager.py # Resource quota management
├── template_system/ # Template system for generating servers
│ ├── template_manager.py # Template loading and parsing
│ ├── customization.py # Customization points
│ ├── handlers/ # Custom handler templates
│ └── templates/ # Server templates
├── client.py # Client for interacting with servers
├── test_protection.py # Security testing tool
├── servers/ # Generated server scripts directory
├── docs/ # Documentation directory
│ ├── api_specification.md # API specification
│ ├── security_protection.md # Security documentation
│ └── development_workflow.md # Development guidelines
└── progress_tracker.py # Development progress tracking utility
API Documentation
For complete API documentation, see the docs/api_specification.md file, which contains detailed information about:
- All available client commands
- Server management endpoints
- Resource monitoring capabilities
- Configuration options
- Auto-scaling functionality
- Security features
- Error handling
Development Status
MCP-Forge development is 100% complete with all 40 planned tasks implemented, including the comprehensive security protection mechanisms.
Run python progress_tracker.py stats to see the overall progress:
Total Tasks: 40
Completed: 40 (100.0%)
In Progress: 0 (0.0%)
Not Started: 0 (0.0%)
Blocked: 0 (0.0%)
Requirements
- Python 3.7+
- MCP SDK 1.6.0+
- httpx 0.28.0+
- anyio 4.0.0+
- psutil 5.9.0+
- requests 2.31.0+
- python-dateutil 2.8.2+
- aiosmtplib 2.0.1+ (for email notifications)
- prometheus-client 0.17.1+ (optional, for exposing metrics)
- jsonschema 4.17.3+ (for configuration validation)
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
MCP Package Docs Server
Facilitates LLMs to efficiently access and fetch structured documentation for packages in Go, Python, and NPM, enhancing software development with multi-language support and performance optimization.
Claude Code MCP
An implementation of Claude Code as a Model Context Protocol server that enables using Claude's software engineering capabilities (code generation, editing, reviewing, and file operations) through the standardized MCP interface.
@kazuph/mcp-taskmanager
Model Context Protocol server for Task Management. This allows Claude Desktop (or any MCP client) to manage and execute tasks in a queue-based system.
Linear MCP Server
Enables interaction with Linear's API for managing issues, teams, and projects programmatically through the Model Context Protocol.
mermaid-mcp-server
A Model Context Protocol (MCP) server that converts Mermaid diagrams to PNG images.
Jira-Context-MCP
MCP server to provide Jira Tickets information to AI coding agents like Cursor
Linear MCP Server
A Model Context Protocol server that integrates with Linear's issue tracking system, allowing LLMs to create, update, search, and comment on Linear issues through natural language interactions.
Sequential Thinking MCP Server
This server facilitates structured problem-solving by breaking down complex issues into sequential steps, supporting revisions, and enabling multiple solution paths through full MCP integration.