AI tooling

dqx ships an MCP server, so an AI agent can ask about your environment instead of guessing at it.

claude mcp add dqx --scope user -- dqx mcp

One registration covers every project on the machine.

What it answers

The split is worth understanding, because it decides which tool an agent should reach for.

dqx knows where. Which projects exist, where a site's code lives on the host and inside the container, what to run and in which directory, where the database is today, who holds 80 and 443, what is broken.

A framework's own tooling knows what. Your routes, your models, your config keys. Laravel Boost does this well, and dqx does not reimplement it — the two sit beside each other, and dqx's tools are prefixed dqx_ so they cannot be confused.

Why it matters more than it sounds

An agent without this guesses, and its guesses look plausible:

• It runs the test suite directly and reports a failure that is not real, because the project declares a prepare-test step it did not know about.
• It reads a docker exec -w /var/www/… line from a stale guideline, lands in a directory that still exists but is half-empty, and concludes a command is missing.
• It writes a database port into a file, and the port changes at the next start.

Each of those is a wrong answer produced confidently. Asking a tool that reads the running machine removes the guess.

What it will not do

Read-only by default. Anything that changes what is running needs the server started with --allow-write, and some things are not exposed at any flag:

	default	--allow-write	never
reads — info, addresses, logs, doctor, scaffolds	✅
running a command the project declared	✅
running anything else		✅
starting and stopping a project	asks first	✅
creating a new project		✅
installing, switching, cleaning or killing an existing project			✅

Creating a project is allowed and installing one is not, which sounds inconsistent until you see the line: the tool refuses any directory that already holds a .dq/config.yml, so it can only ever create in a directory it just made. It can never re-install over something somebody is working in.

Without MCP

Every fact is available to a tool that can only run a command:

dqx info --json      # this project: sites, checkouts, container paths, services
dqx list --json      # every project on the machine
dqx doctor --json    # findings with a severity, and a verdict
dqx verify --json    # what every declared domain answered

Each payload carries the dqx version, and the shapes only ever gain fields.

In the project's own guidelines

dqx sync writes an environment note into the project — where its code runs, what to run in it, which services it has — in the file the project's own AI tooling composes, so an agent reads it without being told to.