defanglabs mcp

Defanglabs MCP Server

Defang connects Docker Compose deployment and cloud stack management with AI coding workflows. Developers can prepare AWS, Azure, or GCP stacks, change configuration, deploy applications, inspect services and logs, and remove deployments.

Server facts

defanglabs mcp

Tech

Runtime / language
go
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
Not recordedPart of the defanglabs/defang monorepo
Last commit
Data as of

Tags

  • Aws
  • cloud
  • deployment
  • do
  • gcp

Overview

Docker Compose projects are the starting point for Defang’s deployment workflow, with tools exposed to AI coding agents through MCP. Developers can generate stack files for AWS, Azure, and GCP, inspect available stacks, check the active selection, and choose a deployment target. Configuration tools let them list, set, or remove variables associated with that selection. The deploy tool uses Compose definitions from the working directory to initiate application deployment, while destroy removes the application associated with the selected stack. A separate tool retrieves the project name.

For deployed applications, service inspection provides names, deployment identifiers, public addresses, and status information. Log retrieval supports time boundaries and pagination, allowing developers to examine output across a chosen period. Cost estimation covers AWS and GCP projects. Deployment progress can also be viewed in the Defang Portal. Supported clients include Cursor, Windsurf, Codex, Claude Desktop, and both VS Code editions listed in the documentation: standard and Insiders. The implementation uses Go, communicates over stdio, and carries an MIT license.

defanglabs mcp setup notes

  • defanglabs mcp records stdio only; confirm a network transport upstream before planning remote access.
  • defanglabs mcp: Part of the defanglabs/defang monorepo

Related MCP servers