JFrog Artifactory MCP Server

A Model Context Protocol (MCP) server that provides seamless integration with JFrog Artifactory, enabling AI assistants to manage artifacts, repositories, and perform cleanup operations through natural language interactions.

🌐 Deployment Options

This MCP server can be deployed in two modes:

1. Local Mode (stdio) - Traditional installation where each developer runs the server locally

2. Remote Mode (HTTP) - NEW! Deploy once on a central server, and all developers connect remotely

👉 Want to host this centrally?

• Quick Start (5 min): QUICK_SETUP.md

• Complete Guide: REMOTE_DEPLOYMENT.md

Benefits of remote deployment:

• ✅ No local installation required for developers

• ✅ Single point of updates and maintenance

• ✅ Centralized monitoring and logging

• ✅ Consistent version across team

Related MCP server: JFrog MCP Server

🚀 Features

Artifact Management

• 📄 List Artifacts - Browse repository contents and folder structures

• 📥 Download Artifacts - Retrieve artifacts from repositories to local filesystem

• 📤 Upload Artifacts - Deploy artifacts to repositories

• 🔍 Get Artifact Details - Retrieve metadata, checksums, and properties

• 🗑️ Delete Artifacts - Remove individual artifacts with safety confirmations

Repository Operations

• 📋 List Repositories - Browse available repositories with optional filtering

• 🔧 Repository Management - Access repository configurations and details

Cleanup & Maintenance

• 🧹 Smart Cleanup - Remove artifacts older than specified days/weeks

• 🔍 Search Old Artifacts - Find artifacts by age with detailed size information

• 📊 Storage Analytics - Calculate storage usage and cleanup impact

• 🛡️ Dry Run Mode - Preview cleanup operations before execution

Safety Features

• ⚠️ Confirmation Required - Explicit confirmation needed for deletions

• 🔒 Input Validation - Prevents accidental root deletions and invalid paths

• ✅ Existence Checks - Verifies artifacts exist before operations

• 📝 Detailed Reporting - Comprehensive operation logs and error handling

📋 Requirements

• Python 3.8 or higher

• JFrog Artifactory instance (Cloud or On-Premise)

• Valid JFrog authentication credentials

🛠️ Installation

1. Clone the Repository

git clone https://github.com/your-org/jfrog-mcp-server.git
cd jfrog-mcp-server

2. Install Dependencies

pip install -r requirements.txt

3. Install the Package

pip install -e .

⚙️ Configuration

Environment Variables

Create a .env file in the project root:

# JFrog Artifactory Configuration
JFROG_BASE_URL=https://your-instance.jfrog.io/artifactory
JFROG_ACCESS_TOKEN=your-access-token

# Alternative: Username/Password Authentication
# JFROG_USERNAME=your-username
# JFROG_PASSWORD=your-password

# Optional: Logging Configuration
LOG_LEVEL=INFO

Authentication Options

Option 1: Access Token (Recommended)

JFROG_BASE_URL=https://your-instance.jfrog.io/artifactory
JFROG_ACCESS_TOKEN=your-access-token

Option 2: Username/Password

JFROG_BASE_URL=https://your-instance.jfrog.io/artifactory
JFROG_USERNAME=your-username
JFROG_PASSWORD=your-password

MCP Client Configuration

Add to your MCP client configuration (e.g., Claude Desktop config.json):

{
  "mcpServers": {
    "jfrog-artifactory": {
      "command": "python",
      "args": ["-m", "jfrog_mcp"],
      "env": {
        "JFROG_BASE_URL": "https://your-instance.jfrog.io/artifactory",
        "JFROG_ACCESS_TOKEN": "your-access-token"
      }
    }
  }
}

🌐 Running as Remote HTTP Server

To deploy the MCP server as a centralized HTTP service:

Quick Start with Docker

# Build and run
docker-compose up -d

# Check status
curl http://localhost:8000/health

Run with Python

# Start HTTP server on port 8000
python -m jfrog_mcp --http

# Or specify custom host/port
python -m jfrog_mcp --http --host 0.0.0.0 --port 8000

📚 For complete remote deployment guide, security setup, and client configuration, see REMOTE_DEPLOYMENT.md

🔧 Usage Examples

Repository Management

# List all repositories
get_repositories()

# Filter repositories by type
get_repositories(package_type="maven")

Artifact Operations

# Browse repository contents
list_artifacts("my-repo-local")
list_artifacts("my-repo-local", "path/to/folder")
list_artifacts("my-repo-local", "", deep=True)  # Recursive listing

# Get artifact details
get_artifact_details("libs-release-local", "com/example/app/1.0.0/app-1.0.0.jar")

# Download an artifact
download_artifact("libs-release-local", "path/to/artifact.jar", "/local/path/artifact.jar")

# Upload an artifact
push_artifact("libs-release-local", "com/example/app/1.0.0/app-1.0.0.jar", "/local/path/app-1.0.0.jar")

Safe Deletion

# This will show a warning and NOT delete
delete_artifact("my-repo", "path/to/artifact.jar")

# This will actually delete the artifact
delete_artifact("my-repo", "path/to/artifact.jar", confirm_deletion=True)

Cleanup Operations

# Search for old artifacts (safe preview)
search_old_artifacts("libs-snapshot-local", older_than_days=30)

# Preview cleanup (dry run - safe)
cleanup_old_artifacts("libs-snapshot-local", older_than_days=30, dry_run=True)

# Execute cleanup
cleanup_old_artifacts("libs-snapshot-local", older_than_days=30, dry_run=False)

# Target specific folders
cleanup_old_artifacts("docker-local", older_than_days=60, folder_path="old-images/", dry_run=False)

🛠️ Available Tools

🛡️ Safety Features

Deletion Protection

• All deletion operations require explicit confirm_deletion=True parameter

• Input validation prevents empty paths and repository root deletion

• Existence verification before attempting deletion

• Clear error messages for invalid operations

Dry Run Mode

• Cleanup operations default to dry_run=True

• Preview exactly what would be deleted before execution

• Detailed reports showing affected artifacts and storage impact

Error Handling

• Comprehensive error messages with actionable guidance

• Graceful handling of network issues and authentication failures

• Detailed logging for troubleshooting

🐛 Troubleshooting

Common Issues

Authentication Errors

Error: 401 Unauthorized

• Verify your access token or username/password

• Check token permissions in JFrog Artifactory

• Ensure base URL is correct

Connection Issues

Error: Connection timeout

• Verify JFrog Artifactory URL is accessible

• Check network connectivity and firewall settings

• Validate base URL format (should include /artifactory)

Permission Errors

Error: 403 Forbidden

• Verify your user has appropriate repository permissions

• Check if repository exists and is accessible

• Review JFrog permission model for your user/token

Debug Mode

Enable debug logging:

export LOG_LEVEL=DEBUG

🤝 Contributing

1. Fork the repository

2. Create a feature branch (git checkout -b feature/amazing-feature)

3. Commit your changes (git commit -m 'Add amazing feature')

4. Push to the branch (git push origin feature/amazing-feature)

5. Open a Pull Request

Development Setup

# Install development dependencies
pip install -e ".[dev]"

# Run tests
python -m pytest tests/

# Run linting
flake8 jfrog_mcp/

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🆘 Support

For support and questions:

• Open an issue on GitHub

• Check the JFrog Artifactory REST API documentation

• Review MCP protocol documentation

🏗️ Architecture

jfrog_mcp/
├── __init__.py          # Package initialization
├── __main__.py          # CLI entry point
├── config.py            # Configuration management
├── server.py            # MCP server and tool definitions
└── api/
    ├── __init__.py
    └── artifactory.py   # JFrog Artifactory API client

🔮 Future Features

• Build information management

• Repository creation and configuration

• Advanced search with AQL (Artifactory Query Language)

• Artifact properties management

• Replication status monitoring

• Docker registry specific operations

• Maven/Gradle metadata handling

• Bulk operations with progress tracking

Made with ❤️ for the JFrog and MCP communities