Tugboat MCP Server
About
A Model Context Protocol (MCP) server for interacting with the Tugboat API.
Explore
- Access Tugboat projects, previews, and repositories
- Create, build, refresh, and delete previews
- Search for Tugboat resources
- View preview logs
- Support for both stdio and HTTP transports
- Authentication and authorization support
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Tugboat MCP ServerCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
npm install
The following environment variables are required:
- TUGBOAT_API_KEY: Your Tugboat API key
- TRANSPORT_TYPE: The transport type to use (stdio or http, defaults to stdio)
- PORT: The port to use for HTTP transport (defaults to 3000)
- TUGBOAT_API_URL: The base URL for the Tugboat API (defaults to https://api.tugboatqa.com/v3)
When using Claude Desktop, authentication is handled automatically through the TUGBOAT_API_KEY environment variable that you provide in the configuration. The stdio transport used by Claude Desktop doesn't require the explicit authentication steps that the HTTP transport does.
Like Claude Desktop, Cursor handles authentication automatically through the environment variables provided in the configuration. The MCP server uses the TUGBOAT_API_KEY to authenticate with the Tugboat API.
When using the HTTP transport, you need to authenticate explicitly:
1. Get an authentication token:
curl -X POST http://localhost:3000/auth/login
Response:
{
"success": true,
"token": "your-tugboat-api-key"
}
2. Use the token to access the MCP endpoint:
curl -X POST http://localhost:3000/mcp \
-H "Authorization: Bearer your-tugboat-api-key" \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","method":"initialize","params":{},"id":1}'
To run the tests, use the following command:
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"tugboat mcp server": {
"tugboat-mcp": {
"command": "node",
"args": [
"/path/to/tugboat-mcp/dist/index.js"
],
"env": {
"TUGBOAT_API_KEY": ""
}
}
}
}
}McpServers
{
"tugboat-mcp": {
"command": "node",
"args": [
"/path/to/tugboat-mcp/dist/index.js"
],
"env": {
"TUGBOAT_API_KEY": ""
}
}
}A Model Context Protocol (MCP) server for interacting with the Tugboat API. This server allows AI assistants like Claude to access and manipulate Tugboat resources through the standardized MCP interface.
What is MCP?
The Model Context Protocol (MCP) is an open protocol created by Anthropic that enables seamless integration between AI assistants and external data sources or tools. It provides a standardized way for AI models to:
- Access resources (data and context)
- Use tools (functions to perform actions)
- Follow prompts (templated workflows)
This Tugboat MCP server implements the protocol to expose Tugboat's API capabilities to AI assistants like Claude.
Features
- Access Tugboat projects, previews, and repositories
- Create, build, refresh, and delete previews
- Search for Tugboat resources
- View preview logs
- Support for both stdio and HTTP transports
- Authentication and authorization support
Architecture
The server follows a modular architecture:
- Core: Main server setup and configuration management
- Resources: Exposes Tugboat entities as MCP resources
- Tools: Implements functions to interact with Tugboat API
- Utils: API client and configuration utilities
- Auth: Authentication and authorization management
- Middleware: HTTP request handling for authentication
Installation
```bash
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
