Skip to main content
MCP server CX Codex

Antideploy — Claude Code MCP server: deploy from your agent

An MCP server that lets your coding agent ship apps to Antideploy. You approve one link, the agent handles the rest.

Install into Claude Code · Platform CLI All methods and platforms ↓
claude mcp add --env ANTIDEPLOY_API_KEY=<ANTIDEPLOY_API_KEY> --env ANTIDEPLOY_URL=<ANTIDEPLOY_URL> --transport stdio antideploy -- npx -y antideploy-mcp
Open on GitHub

What Antideploy does

Antideploy plugs the Antideploy platform into your coding agent over MCP. You say what you want built, the agent writes it and calls login, which returns a link and a short code. Open the link, check the code, hit Approve — that is your only manual step. From there the agent registers the application, spins up a database if the project needs one, and calls deploy.

The server exposes a set of tools over the platform's HTTP API: create_application, deploy, deployment_status, rollback, logs, health, metrics, plus environment variables, databases, storage, sign-in and email for the app. Deploys run asynchronously — deploy, redeploy and rollback return a taskId, and you poll deployment_status until it reads succeeded or failed. Each call already waits up to 20 seconds, so no sleep between polls.

Secrets stay out of the conversation. login never prints the token, and the *_env_to_file tools write values straight into a .env* file and reply with variable names only. A .env file is not uploaded, nor are node_modules, build output or private keys. A folder is bound to its application through .antideploy.json; a folder without one gets refused rather than matched by guesswork. Watch the limits: 4,000 files and 28 MB per project, 28 MB for a single file, one deploy per application at a time. Bigger projects go through GitHub. Deleting apps, custom domains and billing stay in the console on purpose.

What it helps with

  • ship an agent-written app without a Dockerfile or YAML
  • provision a database and write its variables into .env
  • read deploy logs and status from inside the agent
  • roll back a bad release to the previous version
  • schedule a recurring task inside the app

Example prompts for the agent

  • Deploy this project on Antideploy
  • Show me the logs from the last deploy and check health
  • Create a database for the app and dump the variables into .env

How to install Antideploy

Platform CLI

claude mcp add --env ANTIDEPLOY_API_KEY=<ANTIDEPLOY_API_KEY> --env ANTIDEPLOY_URL=<ANTIDEPLOY_URL> --transport stdio antideploy -- npx -y antideploy-mcp

Replace <…> with your own values

Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more

Questions about Antideploy

How do I install Antideploy in Claude Code?

Through the plugin marketplace: run claude plugin marketplace add antideploy/antideploy-mcp, then claude plugin install antideploy@antideploy. The plugin also brings the skills/antideploy skill.

What does Antideploy do?

It hands your agent tools to deploy on the Antideploy platform: create an app, a database, env vars, logs, metrics, rollback. You confirm sign-in through one link and the agent does the rest.

Do I need an API key?

No — sign-in goes through login and the token is kept in ~/.antideploy/config.json. ANTIDEPLOY_API_KEY is for a pipeline or a second machine: a server started with it is bound to one application and skips login.

Similar MCP servers

All MCP servers for DevOps and infrastructure →

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor