Doppler MCP Server
A Model Context Protocol (MCP) server that wraps the Doppler CLI for seamless secrets management integration with Claude and other MCP clients.
Overview
This MCP server provides a bridge between MCP-compatible applications (like Claude Desktop) and the Doppler CLI, allowing you to manage secrets, projects, configs, and environments through natural language interactions.
Prerequisites
- Node.js 18 or higher
- Doppler CLI installed and authenticated
- An active Doppler account with appropriate permissions
Installation
From Source```bash
Clone or navigate to the repository
cd doppler-mcp-server
Install dependencies
npm install
Build the project
npm run build
## Configuration
### Claude Desktop
Add this server to your Claude Desktop configuration file:
**MacOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`
**Windows**: `%APPDATA%\Claude\claude_desktop_config.json````json
{
"mcpServers": {
"doppler": {
"command": "node",
"args": ["/absolute/path/to/doppler-mcp-server/build/index.js"]
}
}
}
Other MCP Clients
For other MCP clients, configure them to run:```bash
node /path/to/doppler-mcp-server/build/index.js
## Available Tools
The server exposes the following Doppler CLI operations as MCP tools:
### Secrets Management
- **doppler_secrets_get** - Get a specific secret value
- Parameters: `name` (required), `project`, `config`
- **doppler_secrets_list** - List all secrets in a config
- Parameters: `project`, `config`
- **doppler_secrets_set** - Set a secret value
- Parameters: `name` (required), `value` (required), `project`, `config`
- **doppler_secrets_delete** - Delete a secret
- Parameters: `name` (required), `project`, `config`
### Project Management
- **doppler_projects_list** - List all projects
- Parameters: none
- **doppler_projects_create** - Create a new project
- Parameters: `name` (required), `description`
### Config Management
- **doppler_configs_list** - List all configs in a project
- Parameters: `project`
- **doppler_configs_create** - Create a new config
- Parameters: `name` (required), `project` (required), `environment` (required)
### Environment Management
- **doppler_environments_list** - List all environments in a project
- Parameters: `project`
### Utility
- **doppler_run** - Run a command with Doppler secrets injected
- Parameters: `command` (required), `project`, `config`
- **doppler_me** - Get information about the authenticated user
- Parameters: none
## Usage Examples
Once configured, you can interact with Doppler through your MCP client:```
"List all my Doppler projects"
"Get the API_KEY secret from my production config"
"Set DATABASE_URL to postgres://... in the dev config"
"Create a new project called my-app"
"Show me all secrets in the staging config"
Authentication
This server uses the Doppler CLI's authentication. Make sure you've authenticated the Doppler CLI before using this server:```bash
doppler login
You can verify your authentication status:```bash
doppler me
Development
Build```bash
npm run build
### Watch Mode```bash
npm run watch
How It Works
This server:
- Listens for MCP tool requests via stdio
- Translates tool calls into Doppler CLI commands
- Executes the commands using your local Doppler CLI installation
- Returns the results in JSON format back to the MCP client
Troubleshooting
"Command not found: doppler"
Make sure the Doppler CLI is installed and available in your PATH:```bash
which doppler
doppler --version
### "Authentication required"
Authenticate with Doppler:```bash
doppler login
Server not appearing in Claude Desktop
- Check that the path in
claude_desktop_config.jsonis correct and absolute - Restart Claude Desktop completely
- Check the Claude Desktop logs for errors
Security Considerations
- This server executes Doppler CLI commands with the permissions of the authenticated user
- Secrets are transmitted through the MCP protocol - ensure your MCP client is trusted
- The server does not store any secrets, it only proxies requests to the Doppler CLI
License
MIT
Contributing
Contributions are welcome! Please feel free to submit issues or pull requests.
