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

# Quickstart

> Install wiggum and ship your first feature in 5 minutes.

## Prerequisites

* Node.js 18 or later
* An API key from [Anthropic](https://console.anthropic.com/), [OpenAI](https://platform.openai.com/), or [OpenRouter](https://openrouter.ai/)
* [Claude Code](https://docs.anthropic.com/en/docs/claude-code) or another CLI-based coding agent (for the autonomous loop)

## Install

<CodeGroup>
  ```bash npm theme={null}
  npm install -g wiggum-cli
  ```

  ```bash npx theme={null}
  npx wiggum-cli
  ```
</CodeGroup>

## Ship a feature

<Steps>
  <Step title="Launch wiggum">
    Open the interactive TUI by running `wiggum` with no arguments:

    ```bash theme={null}
    wiggum
    ```
  </Step>

  <Step title="Scan your project">
    Run `/init` in the TUI. Wiggum detects your tech stack and generates configuration in a `.ralph/` directory.

    ```
    /init
    ```

    You'll be prompted to select your AI provider and enter your API key.
  </Step>

  <Step title="Create a feature spec">
    Run `/new` with a feature name. Wiggum starts an AI-guided interview to understand what you want to build.

    ```
    /new add-dark-mode
    ```

    Answer 3-5 questions about your feature. Wiggum generates a detailed spec in `.ralph/specs/`.
  </Step>

  <Step title="Run the autonomous loop">
    Run `/run` with the feature name. Wiggum delegates to your coding agent and runs implement → test → fix cycles.

    ```
    /run add-dark-mode
    ```

    The loop runs until all tests pass or it reaches the maximum iterations.
  </Step>
</Steps>

## What's next

* Learn about [interactive mode](/using-wiggum/interactive-mode) and all available TUI commands
* Configure [AI providers](/using-wiggum/configuration) and customize loop behavior
* Read the full [command reference](/commands/init)