> ## Documentation Index
> Fetch the complete documentation index at: https://docs.pre.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Coding Agent

> Describe an app. pre.dev builds it, ships real pull requests to your GitHub, and deploys.

The **Coding Agent** is pre.dev's flagship. Point it at a prompt, a template, or an existing GitHub repo — it plans the architecture, writes the code, opens real PRs, and deploys the result. You stay in the loop as a reviewer, not a typist.

<Card title="Build your first project" icon="play" href="/coding-agent/quickstart" horizontal>
  Sign up, describe an app, hit **Build on Autopilot**. \~5 minutes to a deployed product.
</Card>

## How it works

<CardGroup cols={3}>
  <Card title="1. Plan" icon="compass" href="/coding-agent/specifications/fast-vs-deep">
    One prompt → a full structured spec: stack, milestones, user stories, acceptance criteria. Pick Fast (\~30s) or Deep (\~2–5min).
  </Card>

  <Card title="2. Build" icon="hammer" href="/coding-agent/building/build-modes">
    Focused agents work the roadmap in isolated sandboxes. Each task verifies itself before it ships.
  </Card>

  <Card title="3. Ship" icon="rocket" href="/coding-agent/building/pull-requests">
    Real PRs land on feature branches in your GitHub. Review, merge, deploy. Custom domains supported.
  </Card>
</CardGroup>

## What you get

<CardGroup cols={2}>
  <Card title="Working integrations, not stubs" icon="plug">
    Stripe, Supabase, Clerk, OpenAI, Linear, Slack — any service you've connected. Real keys, real endpoints, real data. No `YOUR_API_KEY` placeholders.
  </Card>

  <Card title="Verified code, not vibes" icon="shield-check">
    Every PR passes types, lint, tests, and browser verification before it's opened.
  </Card>

  <Card title="Progress you can watch" icon="eye">
    Live timeline of the agent's work — which file it's editing, which test just passed, which PR just opened.
  </Card>

  <Card title="Your choice of control" icon="sliders">
    Autopilot (ship the whole roadmap), one task at a time (review each PR), or chat (interactive).
  </Card>
</CardGroup>

## Extend what the agent can do

Agents are only as capable as the services they can reach. Connect your stack once — every project inherits it.

<CardGroup cols={2}>
  <Card title="OAuth Connectors" icon="plug" href="/coding-agent/integrations/oauth">
    Sign in once (GitHub, Linear, Slack, Notion, etc.) so the agent can read and write on your behalf.
  </Card>

  <Card title="API Keys & Secrets" icon="key" href="/coding-agent/integrations/api-keys">
    Store provider keys (Stripe, OpenAI, Twilio) once and inject them into every build.
  </Card>

  <Card title="MCP Servers" icon="server" href="/coding-agent/integrations/mcp-servers">
    Attach any MCP server to expand the tool set the agent can call at build time.
  </Card>

  <Card title="Agent Skills" icon="wand-magic-sparkles" href="/coding-agent/integrations/skills">
    Custom instructions injected into every prompt — your conventions, your constraints, your voice.
  </Card>
</CardGroup>

## Who it's for

<CardGroup cols={2}>
  <Card title="Founders & PMs" icon="lightbulb">
    Ship without writing code. Stay focused on product direction while the agent handles implementation.
  </Card>

  <Card title="Solo developers" icon="user">
    Compress weeks of boilerplate into hours. Spend your time on the parts that actually need you.
  </Card>

  <Card title="Teams" icon="users">
    Queue work as specs and review PRs — the agent handles the rote 70%, you focus on the 30% that matters.
  </Card>

  <Card title="Agencies" icon="building">
    Standardize delivery across client projects. Consistent stacks, consistent quality, faster turnaround.
  </Card>
</CardGroup>