Deploy with AI Agent

Use Claude Code, Cursor, OpenCode, or any AI coding agent to deploy your application on Qovery with a single prompt.

Install the Qovery Agent Skill and let your AI coding agent deploy your application end-to-end — from analyzing your codebase to a running deployment on Kubernetes. No Kubernetes knowledge required.

Skill vs MCP Server — what’s the difference?

• Qovery Agent Skill (this page) = Forward engineering. Takes your source code and deploys it on Qovery. The AI agent analyzes your project, creates Dockerfiles, provisions databases, sets up environment variables, and deploys everything.
• Qovery MCP Server = Operations. Manages existing infrastructure. Query environments, troubleshoot deployments, monitor services.

Both can be used together: use the skill to deploy, then the MCP Server to manage.

---

Install the Skill

One command installs the skill globally for all your projects:

1. Run the installer

   curl -fsSL https://skill.qovery.com/install.sh | bash
   

   This installs the skill to all compatible tool directories automatically. No manual configuration needed.

2. Open your AI coding tool

   Launch Claude Code, Cursor, OpenCode, VS Code Copilot, Gemini CLI, or any compatible tool.

3. Ask the agent to deploy

   Deploy my application with Qovery
   

   The agent loads the skill automatically and guides you through the entire process.

To install for the current project only (instead of globally), run:

curl -fsSL https://skill.qovery.com/install.sh | bash -s -- --project

To update to the latest version, re-run the install command.

---

Compatible Tools

The Qovery Agent Skill follows the Agent Skills open standard and works with 30+ AI coding tools. The installer places the skill in all discovery paths automatically.

Tool	Supported
Claude Code	Yes
Cursor	Yes
OpenCode	Yes
VS Code Copilot	Yes
Gemini CLI	Yes
Roo Code	Yes
Goose	Yes
Amp	Yes
Kiro	Yes
Junie (JetBrains)	Yes
OpenHands	Yes
OpenAI Codex	Yes
Mistral Vibe	Yes
TRAE	Yes

And any other tool that discovers skills from .claude/skills/ or .agents/skills/ directories.

---

What the Skill Does

When you tell your AI agent to deploy, the skill guides it through a complete deployment workflow:

1. Analyzes your codebase — detects language, framework, ports, database needs, environment variables
2. Creates a Dockerfile if one is missing — production-ready multi-stage templates for Node.js, Next.js, React, Vite, Python, Go, Java, Ruby, PHP, .NET
3. Asks the right questions — dev vs production, database type, deployment method
4. Sets up infrastructure — cluster creation from scratch if needed (AWS, GCP, Azure, Scaleway)
5. Deploys via CLI + API (quick path) or Terraform provider (recommended for production)
6. Provisions databases — container mode for dev/test, managed mode (e.g. AWS RDS) for production, or Terraform services for RDS Aurora
7. Sets up environment variables using aliases, interpolation, and overrides — no duplication
8. Handles Helm charts, Terraform modules, lifecycle jobs, and cron jobs
9. Watches deployments and auto-fixes failures — diagnoses build errors, port mismatches, health check failures, missing env vars, and OOM issues. Fixes Qovery configuration automatically; asks for permission before modifying your code

---

What You Need

Qovery account — Sign up at console.qovery.com (free to start)

Git repository connected to Qovery (GitHub, GitLab, or Bitbucket)

That’s it. The skill handles everything else:

• No API token needed upfront — the skill generates one via the Qovery CLI
• No cluster needed upfront — the skill guides you through cluster creation if you don’t have one
• No Dockerfile needed — the skill creates one if missing
• No Kubernetes knowledge needed — the skill configures health checks, ports, deployment stages, and resource limits

---

How It Compares

	Console (Manual)	AI Agent Skill	MCP Server
Purpose	Full control via web UI	Fastest path from code to deployed	Manage existing infrastructure
Creates Dockerfiles	No	Yes, for 12+ frameworks	No
Provisions databases	Manual configuration	Automatic (asks dev vs prod)	No
Sets up env vars	Manual per variable	Automatic with aliases and overrides	No
Deployment method	Console UI	CLI + API or Terraform	N/A
Monitors & fixes failures	Manual log inspection	Automatic diagnosis and auto-fix	Query-based
Best for	Experienced users, fine-tuning	Developers deploying from code	Day-2 operations

---

Supported Frameworks

The skill includes production-ready Dockerfile templates for:

Language	Frameworks
Node.js	Express, Fastify, NestJS
Next.js	SSR with standalone output
React / Vite	SPA served via nginx
Python	Flask, Django, FastAPI
Go	Any (net/http, Gin, Echo, Fiber, etc.)
Java	Spring Boot (Maven and Gradle)
Ruby	Rails
PHP	Laravel
.NET	ASP.NET Core

If your framework is not listed, the agent creates a custom Dockerfile based on your project structure.

---

Example Prompts

Once the skill is installed, try any of these prompts with your AI coding agent:

• “Deploy my application with Qovery”
• “Set up Qovery for my project”
• “Deploy this to Kubernetes with Qovery”
• “Create a Qovery Terraform configuration for my app”
• “I need a production deployment with a PostgreSQL database”

---

Next Steps

Qovery MCP Server

Manage deployed infrastructure from your AI agent

Terraform Provider

Version-controlled infrastructure as code

CLI Reference

Monitor, debug, and manage from the terminal

Skill Source Code

View, contribute, or fork on GitHub