Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/paramveer-cyber/Deployaar/llms.txt

Use this file to discover all available pages before exploring further.

Deployaar takes a one-line feature request and walks it through a complete software delivery pipeline without a human writing the first line of code. An AI agent clarifies intent, generates a structured PRD, breaks the spec into engineering tasks, writes the code inside a sandboxed environment, opens a GitHub pull request, and runs a depth-branched AI code review — all automatically.

Introduction

Learn what Deployaar is, how it works, and who it’s built for.

Quickstart

Set up Deployaar locally and run your first AI-powered feature pipeline.

Architecture

Explore the monorepo layout, four-layer rule, and service topology.

API Reference

Browse the full REST and tRPC API surface with request and response schemas.

How the Pipeline Works

Deployaar turns a plain-English description into shipped code through seven automated steps.
1

Submit a Feature Request

A developer (or end user) submits a one-line feature request through the Deployaar dashboard. The AI clarification gate immediately checks whether it’s a duplicate or an education question and rejects noise before it enters the pipeline.
2

Generate a PRD

The accepted request is expanded into a structured Product Requirements Document — covering problem statement, goals, non-goals, user stories, acceptance criteria, and edge cases.
3

Break Down Tasks

The PRD is decomposed into ordered engineering tasks queued for the coding agent.
4

Run the Coding Agent

An AI agent executes inside an E2B sandbox: it clones your repository, reads files, writes code, runs terminal commands, and commits changes to a feature branch — all checkpointed via Inngest so work survives restarts.
5

Open a GitHub Pull Request

The agent pushes its branch and opens a pull request via your connected GitHub App installation, with the task summary as the PR description.
6

AI Code Review

The pull request is reviewed at your chosen depth — shallow, standard, or deep — against the original PRD. Issues are categorized by severity and tagged as blocking or non-blocking.
7

Merge

Once all blocking issues are resolved, a human approves and merges. The feature request lifecycle is complete.

Explore the Docs

The Pipeline

Deep dive into each step of the automated delivery pipeline.

Coding Agent

How the E2B sandbox, Inngest workflows, and AI tools work together.

PR Review

Depth-branched AI code review — shallow, standard, and deep.

GitHub Integration

Install the GitHub App, link repositories, and manage webhooks.

Billing Plans

Free, Pro, and Pro Max plans with per-plan AI model access.

Environment Variables

Full reference for every environment variable Deployaar requires.

Build docs developers (and LLMs) love