Overview

Introduction

Welcome to the documentation. This guide will help you understand the core concepts, get up and running quickly, and make the most of everything the platform has to offer.

What is this?

This platform provides a powerful set of tools and APIs designed for developers and teams who want to build, deploy, and scale their applications with ease. It is built around simplicity, flexibility, and performance.

Key Features

• Simple integration — connect your existing tools in minutes.

• Scalable infrastructure — built to grow alongside your project.

• Developer-first — clean APIs, great docs, and flexible SDKs.

• Real-time updates — stay in sync without polling.

Getting Started

To get started, head over to the Quickstart guide. It walks you through setting up your environment, making your first API call, and exploring the core features step by step.

If you have questions or run into issues, check the rest of the documentation or reach out to our support team. We are here to help.

What is Queue?

Queue is an AI coding agent platform that lets you build, run, and deploy autonomous agents that understand your codebase. Rather than asking an AI for a code snippet and pasting it in yourself, Queue agents plan, write, test, and ship code on your behalf — working across files, repos, and environments just like a senior engineer on your team.

How Queue is different

Most AI coding tools are completions: you prompt, they respond, you decide what to do with the output. Queue agents are autonomous: they receive a goal, break it into steps, use tools to act on your codebase, verify their own work, and iterate until the task is done.

Queue agents can read and write files, run tests, open pull requests, query your issue tracker, and call external APIs — all in one run, without you sitting at the keyboard.

Core building blocks

Agents

An agent is a configured AI process: a model, a system prompt, a set of tools, and policies for how it should behave. You can spin up agents for specific tasks — a refactor agent, a test-writing agent, a code review agent — or build a general-purpose assistant.

Tools

Tools are the actions an agent can take. Queue ships with built-in tools for reading and writing code, running shell commands, searching the web, calling APIs, and interacting with GitHub. You can also define your own tools with custom logic.

Runs

A run is a single execution of an agent against a goal. Each run produces a full trace: every model call, tool invocation, file change, and decision made along the way. Runs are observable, auditable, and reproducible.

Workspaces

Workspaces hold your agents, API keys, integrations, and usage data. You can invite team members, set permission scopes, and track all runs from the workspace dashboard.

Who is Queue for?

Queue is built for engineering teams that want to move faster without growing headcount. It fits naturally into workflows where repetitive or high-context coding tasks slow people down: dependency upgrades, test coverage, boilerplate generation, migration scripts, bug triage, and code review.

It is also a solid foundation for teams building AI-powered developer tooling — internal bots, CI-integrated review agents, or fully automated code pipelines.

Navigating these docs

If you are just getting started, go to Quickstart — you will have an agent running in under five minutes. If you want to understand how the system works before diving in, read Core Concepts. If you are ready to go deeper, the Agents, SDK, CLI, and MCP sections cover everything in detail.