dokploy
MCP server for the Dokploy API. 196 tools across 23 modules. Your AI agent can now deploy apps, manage databases, configure domains, and handle backups -- without you touching a dashboard.
claude mcp add --transport stdio vcode-sh-dokploy-mcp npx @vibetools/dokploy-mcp \ --env DOKPLOY_URL="https://panel.example.com" \ --env DOKPLOY_API_KEY="your-api-key"
How to use
The dokploy-mcp MCP server provides a centralized interface to interact with the Dokploy API, exposing 196 tools across 23 modules for deploying applications, managing databases, configuring domains, handling backups, and more. Once running, you can point your MCP client at the server to access the full capability set, including tool annotations, type-safe parameter validation, and lazy credential loading. If you have the Dokploy CLI installed and authenticated, or you choose to provide credentials via environment variables, the server will authenticate and begin serving commands to your AI agent through the MCP protocol. The tools are organized by module (e.g., Application, Deployment, Domain, Database, Backup, etc.), and you can rely on the generated docs to understand the parameters and their descriptions in each tool.
How to install
Prerequisites:
- Node.js v22 or newer
- npm (comes with Node)
- Access to a Dokploy panel (DOKPLOY_URL and API key if using environment-based authentication)
Step 1: Install Node.js v22+ if you haven't already.
Step 2: Install the MCP server via npx (no global install required) and run it:
npx @vibetools/dokploy-mcp
Step 3: Start the server with configuration (using environment variables or a local config as shown in the README):
# Example with env vars
export DOKPLOY_URL=https://panel.example.com
export DOKPLOY_API_KEY=your-api-key
npx @vibetools/dokploy-mcp
Step 4: Point your MCP client at the running server (adjust command/args to match your setup). Example using npx approach documented in the repo:
{
"mcpServers": {
"dokploy": {
"command": "npx",
"args": ["@vibetools/dokploy-mcp"],
"env": {
"DOKPLOY_URL": "https://panel.example.com",
"DOKPLOY_API_KEY": "your-api-key"
}
}
}
}
Step 5: Optional interactive setup wizard (to save credentials to disk) can be run via:
npx @vibetools/dokploy-mcp setup
This wizard validates credentials, saves them to a config file, and outputs a minimal MCP config to copy. After setup, the env block can be omitted if credentials are stored locally.
Additional notes
Notes and tips:
- The MCP server supports three config resolution sources: environment variables, a local config file (e.g., ~/.config/dokploy-mcp/config.json), and the Dokploy CLI config. First match wins.
- The DOKPLOY_TIMEOUT environment variable can override the default request timeout (ms).
- If you prefer not to store credentials in environment variables, run the setup wizard to save credentials to disk and rely on the generated MCP config.
- You can also use the Dokploy CLI auto-detection to skip config entirely if the CLI is authenticated.
- For development or testing, you can point your MCP client at a local build of the server (see the Development section in the README).
Related MCP Servers
mcp-security-hub
A growing collection of MCP servers bringing offensive security tools to AI assistants. Nmap, Ghidra, Nuclei, SQLMap, Hashcat and more.
coolify
MCP server for Coolify — 38 optimized tools for managing self-hosted PaaS through AI assistants
aws
A lightweight service that enables AI assistants to execute AWS CLI commands (in safe containerized environment) through the Model Context Protocol (MCP). Bridges Claude, Cursor, and other MCP-aware AI tools with AWS CLI for enhanced cloud infrastructure management.
agentql
Model Context Protocol server that integrates AgentQL's data extraction capabilities.
workflowy
Powerful CLI and MCP server for WorkFlowy: reports, search/replace, backup support, and AI integration (Claude, LLMs)
opnsense
Modular MCP server for OPNsense firewall management - 88 tools providing access to 2000+ methods through AI assistants