GETTING STARTED

Install QUIQQER with AI, Docker or PHAR

Start with an AI agent, explore a ready-to-use Docker demo, or install a persistent QUIQQER system on your server.

Start with your AI agent Choose an installation method

choose your path

FASTEST WAY TO START

Start immediately with your AI agent.

Give your AI assistant access to QUIQQER’s official setup instructions. This connection is called MCP — it lets your assistant look up the right steps while helping you install QUIQQER.

  1. 1
    Connect your AI assistant

    Choose the app you use. Have it installed and signed in, and open an empty folder for your new website.

CodexClaude CodeCursorVS Code / Copilot

Connect Codex

Using the Codex CLI? Open a terminal (the app where you type commands) and run:

codex mcp add quiqqer --url https://quiqqer.com/mcp

Open your project folder in Codex and start a new session. Type /mcp to check that QUIQQER is connected.

Using the Codex app or IDE extension?

Add this to your Codex configuration file, ~/.codex/config.toml, then restart Codex. If a [mcp_servers.quiqqer] entry already exists, update it.

[mcp_servers.quiqqer]
url = "https://quiqqer.com/mcp"
Official Codex connection guide ↗

Connect Claude Code

Open a terminal in the folder where you want to create your website and run:

claude mcp add --transport http quiqqer https://quiqqer.com/mcp

Start Claude Code with claude. Type /mcp and check that QUIQQER is connected.

This example uses Claude Code, which can create files and run the installation on your computer.

Official Claude Code connection guide ↗

Connect Cursor

  1. Open your project folder in Cursor.
  2. Create a folder called .cursor and a file inside it called mcp.json. Paste the configuration below. If the file already has servers, add the quiqqer entry to its existing mcpServers object.
  3. Save the file, enable QUIQQER in Cursor’s MCP settings, then open an Agent chat.
{
  "mcpServers": {
    "quiqqer": {
      "url": "https://quiqqer.com/mcp"
    }
  }
}
Official Cursor connection guide ↗

Connect GitHub Copilot in VS Code

  1. Open your project folder in VS Code with GitHub Copilot set up.
  2. Open the Command Palette: Ctrl + Shift + P (Mac: ⌘ + Shift + P).
  3. Choose MCP: Add Server, then HTTP. Paste this address:
https://quiqqer.com/mcp

Name the server quiqqer and save it to your workspace. Use MCP: List Servers to start it, review the trust prompt, then open Copilot Chat in Agent mode.

Official VS Code connection guide ↗

Connected? Continue with the starter prompt below. If QUIQQER does not appear, check the server address and restart your AI app.

  1. 2
    Copy the starter prompt

    Paste the prompt into your connected assistant. Replace the text in square brackets with what you want to build, then send it.

  2. 3
    Approve required access

    The agent asks only when it needs information or permission it cannot determine itself.

Learn how QUIQQER works with AI agents
Starter prompt
Use the connected QUIQQER server (https://quiqqer.com/mcp) and load the quiqqer_start skill.

Set up a new QUIQQER system in the current directory. Choose the most suitable installation method, database and secure defaults for my environment.

My goal is:
[Describe your website or application here]

Ask only for information or permissions that you cannot determine yourself.

TRY QUIQQER

A working demo in one command.

Each demo contains QUIQQER, example data, Apache and an embedded MariaDB database. Choose one image and open it locally.

  • Docker installed and running
  • Port 8080 available
  • No configuration required

Choose one demo to run locally.

Website demo

docker run --detach \
  --name quiqqer-demo \
  --publish 8080:80 \
  quiqqer/demo:website

Shop demo

docker run --detach \
  --name quiqqer-demo \
  --publish 8080:80 \
  quiqqer/demo:shop

ERP demo

docker run --detach \
  --name quiqqer-demo \
  --publish 8080:80 \
  quiqqer/demo:erp

Demo only. Do not expose this container to the internet or use it in production. Its data is not stored independently of the container.

Open the complete demo guide

DOCKER PRODUCTION

Persistent, HTTPS-ready and built to operate.

The recommended Compose setup combines QUIQQER with MariaDB and Caddy. Named volumes preserve application files, database data and certificates.

  • Docker and Docker Compose
  • A public hostname with working DNS
  • Inbound ports 80 and 443
  • Backups and monitoring before launch
QUIQQER · PHP 8.5 · ApacheMariaDB 11.8Caddy 2 · HTTPS

1. Create the environment

mkdir quiqqer-production
cd quiqqer-production

Create .env with your public hostname and a new random database password:

QUIQQER_HOSTNAME=www.example.com
QUIQQER_DB_PASSWORD=replace-with-a-long-random-password

2. Add the Compose configuration

Show the production compose.yaml
name: quiqqer-production

services:
  database:
    image: mariadb:11.8
    restart: unless-stopped
    environment:
      MARIADB_RANDOM_ROOT_PASSWORD: "1"
      MARIADB_DATABASE: quiqqer
      MARIADB_USER: quiqqer
      MARIADB_PASSWORD: ${QUIQQER_DB_PASSWORD}
    volumes:
      - database-data:/var/lib/mysql
    healthcheck:
      test: ["CMD", "healthcheck.sh", "--connect", "--innodb_initialized"]
      start_period: 10s
      interval: 10s
      timeout: 5s
      retries: 10

  quiqqer:
    image: quiqqer/production:php8.5-apache
    restart: unless-stopped
    depends_on:
      database:
        condition: service_healthy
    expose:
      - "80"
    volumes:
      - quiqqer-html:/var/www/html

  caddy:
    image: caddy:2
    restart: unless-stopped
    depends_on:
      - quiqqer
    command: ["caddy", "reverse-proxy", "--from", "${QUIQQER_HOSTNAME}", "--to", "quiqqer:80"]
    ports:
      - "80:80"
      - "443:443"
    volumes:
      - caddy-data:/data
      - caddy-config:/config

volumes:
  database-data:
  quiqqer-html:
  caddy-data:
  caddy-config:

3. Validate and start

docker compose config --quiet
docker compose up --detach
docker compose ps

4. Complete the browser installer

Open your HTTPS hostname. Use database as the database host, port 3306, database and user quiqqer, and the password from .env.

Keep your data. Back up the database and QUIQQER files before updates. Never use docker compose down --volumes during routine operation.

Open the complete production guide

PHAR INSTALLER

The guided CLI installation.

The official installer checks your system and guides you through language, directory, QUIQQER version, template, database and administrator setup.

  • PHP 8.2 or newer
  • PDO and the selected database driver
  • CLI access and a writable installation directory
  • MySQL/MariaDB, PostgreSQL or SQLite

Download and start

wget https://update.quiqqer.com/quiqqer-installer.phar
php quiqqer-installer.phar
System checkDirectoryDatabaseAdmin userFirst project

Automation

For non-interactive installations, pass passwords through protected environment variables instead of command-line arguments:

QUIQQER_DB_PASSWORD='database-password' \
QUIQQER_ADMIN_PASSWORD='administrator-password' \
php quiqqer-installer.phar --no-interaction
Download installerInstaller source & documentation

ADVANCED PATHS

Build, test and adapt.

Official images also support project-specific runtimes and automated package testing.

Custom Docker imagesExtend production, base or base-with-db.Package testingUse dedicated images for CI and ERP tests.TroubleshootingPorts, containers, databases and safe recovery.

NEXT STEPS

Your platform is ready. Start building.

Create your first project, choose a template and extend QUIQQER with the packages your application needs.

Explore extensions Open documentation

AI & MCP

Ready to connect your AI agent?

Connect AI clients to public QUIQQER skills, installation guidance and the capability catalog; machine-readable setup instructions are available in llms.txt.

https://www.quiqqer.com/mcp