MCP Harbor
MCP Harbor is a Node.js application that provides a Model Context Protocol (MCP) server for interacting with Harbor container registry.
README
MCP Harbor
MCP Harbor is a Node.js application that provides a Model Context Protocol (MCP) server for interacting with Harbor container registry.
Table of Contents
Features
- MCP Server: Exposes tools for interacting with Harbor through the Model Context Protocol
- Harbor Operations: Supports operations for projects, repositories, tags, and Helm charts
- TypeScript: Written in TypeScript for better type safety and developer experience
- Automated Tests: Comprehensive test suite for reliable functionality
Prerequisites
Before installing MCP Harbor, ensure you have:
- Node.js 18.x or higher
- npm 8.x or higher
- Access to a Harbor registry instance
- Git (for cloning the repository)
Installation
-
Clone the repository:
git clone https://github.com/nomagicln/mcp-harbor.git -
Navigate to the project directory:
cd mcp-harbor -
Install dependencies:
npm install -
Build the project:
npm run build
Usage
Command Line Arguments
The application accepts the following command line arguments:
Options:
--url Harbor API URL [string] [required]
--username Harbor username (or robot account name) [string] [required]
--password Harbor password (mutually exclusive with --token) [string]
--token Harbor token for robot/service accounts [string]
--insecure Disable TLS/SSL certificate validation [boolean] [default: false]
--debug Enable debug mode [boolean] [default: false]
--help Show help [boolean]
Note: You must provide either
--passwordor--token, but not both.
Environment Variables
Instead of command line arguments, you can also use environment variables. Create a .env file in the root directory:
# Harbor API Configuration
HARBOR_URL=https://harbor.example.com
HARBOR_USERNAME=admin
# Authentication: use either PASSWORD or TOKEN (mutually exclusive)
# Option 1: Password-based authentication
HARBOR_PASSWORD=Harbor12345
# Option 2: Token-based authentication (for robot/service accounts)
# HARBOR_USERNAME=robot$myrobot
# HARBOR_TOKEN=your-robot-token-here
# Debug Mode (true/false)
DEBUG=false
# Disable TLS/SSL certificate validation (true/false)
# HARBOR_INSECURE=false
Token Authentication (Robot/Service Accounts)
Harbor supports robot accounts for automated access. To use token-based authentication:
- Create a robot account in your Harbor instance (Administration > Robot Accounts)
- Use the robot account name as
--username(e.g.,robot$myrobot) - Use the generated token as
--token
mcp-harbor --url https://harbor.example.com --username "robot\$myrobot" --token "your-robot-token"
MCP Client Configuration
Claude Desktop
Add the following to your Claude Desktop configuration file (claude_desktop_config.json):
Using password authentication:
{
"mcpServers": {
"harbor": {
"command": "npx",
"args": [
"mcp-harbor",
"--url", "https://harbor.example.com",
"--username", "admin",
"--password", "your-password"
]
}
}
}
Using token authentication (robot/service account):
{
"mcpServers": {
"harbor": {
"command": "npx",
"args": [
"mcp-harbor",
"--url", "https://harbor.example.com",
"--username", "robot$myrobot",
"--token", "your-robot-token"
]
}
}
}
With self-signed certificates:
{
"mcpServers": {
"harbor": {
"command": "npx",
"args": [
"mcp-harbor",
"--url", "https://harbor.example.com",
"--username", "admin",
"--password", "your-password",
"--insecure"
]
}
}
}
Cursor / Windsurf
Add the following to your MCP configuration file (.cursor/mcp.json or .windsurf/mcp.json):
{
"mcpServers": {
"harbor": {
"command": "npx",
"args": [
"mcp-harbor",
"--url", "https://harbor.example.com",
"--username", "admin",
"--password", "your-password"
]
}
}
}
Using environment variables
You can also reference environment variables to avoid hardcoding credentials:
{
"mcpServers": {
"harbor": {
"command": "npx",
"args": ["mcp-harbor"],
"env": {
"HARBOR_URL": "https://harbor.example.com",
"HARBOR_USERNAME": "admin",
"HARBOR_PASSWORD": "your-password"
}
}
}
}
SSE Transport (remote/shared server)
If you prefer running the MCP server as a standalone process with SSE transport:
npx mcp-harbor --url https://harbor.example.com --username admin --password your-password --sse --port 3000
Then configure your MCP client to connect via SSE at http://localhost:3000/sse.
MCP Tools
The MCP server exposes the following tools:
| Tool Name | Description | Parameters |
|---|---|---|
list_projects |
List all projects in Harbor | None |
get_project |
Get project details by ID | projectId: string |
create_project |
Create a new project | project_name: string, metadata?: object |
delete_project |
Delete a project | projectId: string |
list_repositories |
List repositories in a project | projectId: string |
delete_repository |
Delete a repository | projectId: string, repositoryName: string |
list_tags |
List tags in a repository | projectId: string, repositoryName: string |
delete_tag |
Delete a tag | projectId: string, repositoryName: string, tag: string |
list_charts |
List Helm charts | projectId: string |
list_chart_versions |
List chart versions | projectId: string, chartName: string |
delete_chart |
Delete chart version | projectId: string, chartName: string, version: string |
Development
Running in Development Mode
npm run dev
Running Tests
# Run all tests
npm test
# Run tests with coverage
npm run test:coverage
Project Structure
mcp-harbor
├── src
│ ├── app.ts # Main application entry point (MCP server)
│ ├── definitions
│ │ └── tool.definitions.ts # Tool definitions for MCP
│ ├── services
│ │ └── harbor.service.ts # Harbor service implementation
│ └── types
│ └── index.ts # TypeScript type definitions
├── test
│ └── harbor.test.ts # Tests for Harbor service
├── .env.example # Example environment variables
├── .gitignore # Git ignore file
├── .eslintrc.json # ESLint configuration
├── package.json # Project dependencies
├── jest.config.js # Jest configuration
├── tsconfig.test.json # TypeScript configuration for tests
├── tsconfig.json # TypeScript configuration
├── LICENSE # Project license
└── README.md # Project documentation
Troubleshooting
Common Issues
-
Connection Failed
Error: Unable to connect to Harbor instance- Verify HARBOR_URL is correct and accessible
- Check network connectivity
- Ensure Harbor instance is running
-
Authentication Failed
Error: Invalid credentials- Verify HARBOR_USERNAME and HARBOR_PASSWORD are correct
- Check if user has required permissions
-
Build Errors
Error: TypeScript compilation failed- Run
npm installto ensure all dependencies are installed - Check TypeScript version compatibility
- Clear the
distdirectory and rebuild
- Run
Debug Mode
Enable debug mode by using the --debug flag or setting:
DEBUG=true
Support
For additional help:
- Review the application logs
License
This project is licensed under the MIT License - see the LICENSE file for details.
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.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.