GitHub - alexpota/deploy-mcp: Universal deployment tracker for AI assistants - check deployment status without leaving your AI chat (original) (raw)

deploy-mcp

Universal Deployment Tracker for AI Assistants

Track deployments across all platforms directly in your AI conversation.
No more context-switching. No more dashboard hunting.

Demo

Quick Start

─────────────────────────

Get started in under 30 seconds:

That's it! The server is now running and ready to be configured in your AI assistant.

Supported Platforms

─────────────────────────

deploy-mcp supports multiple deployment platforms simultaneously. Configure as many as you need:

Multi-Platform Configuration

You can use multiple platforms simultaneously by providing tokens for each platform you want to track:

{ "mcpServers": { "deploy-mcp": { "command": "npx", "args": ["-y", "deploy-mcp"], "env": { "VERCEL_TOKEN": "your-vercel-token", "NETLIFY_TOKEN": "your-netlify-token", "CLOUDFLARE_TOKEN": "accountId:globalApiKey" // Add more platform tokens as needed } } } }

Platform Configuration

─────────────────────────

Vercel

Setup Instructions

1. Get your API token:
   • Go to vercel.com/account/tokens
   • Click "Create Token"
   • Give it a descriptive name (e.g., "deploy-mcp")
   • Copy the token
2. Add to your AI assistant configuration:
   {
   "env": {
   "VERCEL_TOKEN": "your-vercel-token"
   }
   }
3. Available commands:
   • "Check my Vercel deployment for project-name"
   • "What's the status of my latest Vercel deployment?"
   • "Show me Vercel deployment logs"
   • "Watch my Vercel deployment progress"
   • "List all my Vercel projects"
   • "Show last 5 deployments for project-name"
4. Required permissions: Read access to deployments and projects

Netlify

Setup Instructions

1. Get your API token:
   • Go to app.netlify.com/user/applications
   • Under "Personal access tokens", click "New access token"
   • Give it a descriptive name (e.g., "deploy-mcp")
   • Copy the token
2. Add to your AI assistant configuration:
   {
   "env": {
   "NETLIFY_TOKEN": "your-netlify-token"
   }
   }
3. Available commands:
   • "Check my Netlify deployment for site-name"
   • "What's the status of my latest Netlify deployment?"
   • "Show me Netlify deployment logs"
   • "Watch my Netlify deployment progress"
   • "List all my Netlify sites"
   • "Show deployment history for site-name"
4. Required permissions: Read access to sites and deploys

Cloudflare Pages

Setup Instructions

1. Get your API token:
   • Go to dash.cloudflare.com/profile/api-tokens
   • Click "Create Token"
   • Use "Custom token" with these permissions:
     * Zone:Zone:Read
     * Zone:Page Rules:Read
     * Account:Cloudflare Pages:Edit
   • Or use your Global API Key (format: accountId:globalApiKey)
   • Copy the token
2. Add to your AI assistant configuration:
   {
   "env": {
   "CLOUDFLARE_TOKEN": "your-cloudflare-token-or-accountId:globalApiKey"
   }
   }
3. Available commands:
   • "Check my Cloudflare Pages deployment for project-name"
   • "What's the status of my latest Cloudflare Pages deployment?"
   • "Show me Cloudflare Pages deployment logs"
   • "Watch my Cloudflare Pages deployment progress"
   • "List all my Cloudflare Pages projects"
   • "Show deployment history for project-name"
4. Token formats supported:
   • API Token: your-api-token (requires CLOUDFLARE_ACCOUNT_ID env var)
   • Global API Key: accountId:globalApiKey (all-in-one format)
5. Required permissions: Account access to Cloudflare Pages

AI Assistant Configuration

─────────────────────────

deploy-mcp works with any MCP-compatible AI assistant. Here's how to set it up:

Claude Desktop

Configuration

File location:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json
• Linux: ~/.config/Claude/claude_desktop_config.json

Configuration:

{ "mcpServers": { "deploy-mcp": { "command": "npx", "args": ["-y", "deploy-mcp"], "env": { "VERCEL_TOKEN": "your-vercel-token", "NETLIFY_TOKEN": "your-netlify-token" } } } }

VS Code

Configuration

File location:

• Workspace: .vscode/mcp.json
• Global: ~/.vscode/mcp.json

Requirements: GitHub Copilot subscription with MCP enabled

Configuration: Same as Claude Desktop

Cursor AI

Configuration

File location: ~/.cursor/mcp.json

Setup: Navigate to Cursor Settings → MCP

Configuration: Same as Claude Desktop

Windsurf IDE

Configuration

File location: ~/.codeium/windsurf/mcp_config.json

Setup: Settings → Advanced Settings → Cascade

Configuration: Same as Claude Desktop

Continue.dev

Configuration

File location: ~/.continue/config.json

Configuration (different structure):

{ "experimental": { "modelContextProtocolServer": { "transport": { "type": "stdio", "command": "npx", "args": ["-y", "deploy-mcp"] }, "env": { "VERCEL_TOKEN": "your-vercel-token", "NETLIFY_TOKEN": "your-netlify-token" } } } }

Cline (VS Code Extension)

Configuration

Natural language setup:

• Ask: "Add the deploy-mcp tool to check deployments"

Manual setup:

• MCP Servers icon → Advanced MCP Settings

Configuration: Same as Claude Desktop

Deployment Status Badges

─────────────────────────

Get live deployment status badges for your repositories:

Badge Format

The deploy-mcp.io endpoint returns JSON data for shields.io. Use the shields.io endpoint wrapper:

Examples by Platform

Vercel Badge

Netlify Badge

Multiple Badges

Webhook Configuration

For real-time badge updates, configure webhooks:

Vercel Webhooks (Pro/Enterprise only)

1. Go to your Vercel team settings
2. Navigate to "Webhooks" section
3. Click "Create Webhook"
4. Configure:
   • URL: https://deploy-mcp.io/webhook/{username}/{repo}/vercel
   • Events: Deployment Created, Ready, Error
   • Projects: Select specific or all Netlify Webhooks
5. Go to Site Settings → Build & deploy → Deploy notifications
6. Add notification → Outgoing webhook
7. Configure:
   • URL: https://deploy-mcp.io/webhook/{username}/{repo}/netlify
   • Events: Deploy started, succeeded, failed

Note: Badges require public repositories. Private repo support coming soon.

Available Tools

─────────────────────────

deploy-mcp provides these tools to your AI assistant:

Core Tools

Platform-Specific Usage

Commands work across all configured platforms:

"Check my Vercel deployment for my-app"
"Show last 10 deployments for my-app on Vercel"
"List all my Vercel projects"
"Check my Netlify deployment for my-site"
"Show me logs for deployment abc123 on Vercel"
"Watch my Netlify deployment progress"
"Show all my Netlify sites"

New Features

List Projects

Discover all your projects/sites across platforms:

"List my Vercel projects"
"Show all Netlify sites"
"What projects do I have on Vercel?"

Deployment History

View multiple recent deployments at once:

"Show last 5 deployments for my-app"
"Get deployment history for my-site"
"Check last 10 deployments on Vercel"

The AI will automatically use the correct platform based on:

1. Explicit platform mention in your command
2. Project/site name matching
3. Available tokens in configuration

Security

─────────────────────────

• Local tokens only - Your API tokens stay on your machine
• No telemetry - We don't track usage or collect data
• Open source - Fully auditable codebase
• Minimal permissions - Only read access required

Advanced Configuration

─────────────────────────

Environment Variables

You can also set tokens via environment variables:

export VERCEL_TOKEN="your-vercel-token" export NETLIFY_TOKEN="your-netlify-token" npx deploy-mcp

Custom Installation

For production use, install globally:

npm install -g deploy-mcp deploy-mcp

Contributing

─────────────────────────

We welcome contributions! See CONTRIBUTING.md for guidelines.

Adding New Platforms

1. Implement the BaseAdapter interface
2. Add platform-specific API client
3. Update configuration documentation
4. Add tests

License

─────────────────────────

Apache 2.0 - See LICENSE for details.

Links

─────────────────────────

• Website
• npm Package
• GitHub Repository
• Report Issues

Made with ❤️ for developers who love AI-assisted coding