inquirer vs prompt-sync vs readline

Building Interactive Command-Line Interfaces in Node.js

inquirer, prompt-sync, and readline are tools for handling user input in Node.js command-line applications. inquirer provides a rich set of interactive UI components like lists and checkboxes using Promises. prompt-sync offers a simple synchronous interface that blocks execution until input is received. readline is a built-in Node.js module that provides low-level stream-based input handling, now enhanced with a Promises API for modern async workflows.

Npm Package Weekly Downloads Trend

3 Years

Github Stars Ranking

Stat Detail

Package	Downloads	Stars	Size	Issues	Publish	License
Package	Downloads	Stars	Size	Issues	Publish	License
inquirer	0	21,491	48.8 kB	24	6 days ago	MIT
prompt-sync	0	226	-	27	6 years ago	MIT
readline	0	101	-	10	10 years ago	BSD

Inquirer vs Prompt-Sync vs Readline: Input Handling Compared

When building command-line tools in Node.js, you need a way to get data from users. inquirer, prompt-sync, and readline all solve this problem but with different trade-offs in complexity, behavior, and dependencies. Let's look at how they handle common tasks.

🖥️ Interaction Model: Rich UI vs Simple Text vs Raw Stream

inquirer provides pre-built interactive components.

You can create lists, checkboxes, and confirmations easily.
It handles arrow keys, filtering, and validation automatically.

// inquirer: Interactive list
import inquirer from 'inquirer';

const { framework } = await inquirer.prompt([
  {
    type: 'list',
    name: 'framework',
    message: 'Choose a framework:',
    choices: ['React', 'Vue', 'Svelte']
  }
]);

prompt-sync captures simple text input.

It shows a message and waits for a line of text.
No built-in menus or selection helpers.

// prompt-sync: Simple text
const prompt = require('prompt-sync')();

const framework = prompt('Choose a framework: ');

readline gives you low-level stream access.

You define the interface and handle the input stream directly.
Requires more code to build menus or validation logic.

// readline: Stream interface
import * as readline from 'node:readline/promises';
import { stdin as input, stdout as output } from 'node:process';

const rl = readline.createInterface({ input, output });
const framework = await rl.question('Choose a framework: ');
rl.close();

⏱️ Execution Flow: Async Promises vs Blocking Sync

inquirer uses Promises and async/await.

It does not block the event loop while waiting for input.
Fits naturally into modern asynchronous Node.js applications.

// inquirer: Async flow
async function start() {
  const { name } = await inquirer.prompt([{ name: 'name', message: 'Name?' }]);
  console.log(`Hello ${name}`);
}

prompt-sync blocks the process synchronously.

Execution stops completely until the user types and hits Enter.
Can cause issues in environments that expect non-blocking I/O.

// prompt-sync: Sync flow
function start() {
  const name = prompt('Name? ');
  console.log(`Hello ${name}`);
}

readline supports both events and Promises.

The modern readline/promises API allows async/await usage.
The older event-based API requires callback management.

// readline: Async flow (Promises API)
async function start() {
  const rl = readline.createInterface({ input, output });
  const name = await rl.question('Name? ');
  rl.close();
  console.log(`Hello ${name}`);
}

📦 Dependencies and Setup: External vs Core

inquirer requires an external npm package.

You must install it via npm install inquirer.
Adds to your dependency tree but saves development time.

# Installation required
npm install inquirer

prompt-sync requires an external npm package.

You must install it via npm install prompt-sync.
Very small footprint but still an external dependency.

# Installation required
npm install prompt-sync

readline is part of Node.js core.

No installation needed — it works out of the box.
Reduces supply chain risk and install time.

// No installation needed
import * as readline from 'node:readline/promises';

✅ Validation and Error Handling

inquirer has built-in validation support.

You can define a validate function in the question config.
It automatically re-prompts the user if input is invalid.

// inquirer: Built-in validation
const { age } = await inquirer.prompt([
  {
    name: 'age',
    message: 'Enter your age:',
    validate: (input) => input > 0 ? true : 'Age must be positive'
  }
]);

prompt-sync has no built-in validation.

You must write manual loops to check and re-ask.
Increases boilerplate code for safe input.

// prompt-sync: Manual validation
let age;
while (!age || age <= 0) {
  age = parseInt(prompt('Enter your age: '));
  if (age <= 0) console.log('Age must be positive');
}

readline has no built-in validation.

You must implement logic to check input and re-prompt.
Gives full control but requires more custom code.

// readline: Manual validation
let age;
while (!age || age <= 0) {
  age = parseInt(await rl.question('Enter your age: '));
  if (age <= 0) console.log('Age must be positive');
}

🌐 Real-World Scenarios

Scenario 1: Scaffolding Tool (e.g., Create-React-App)

You need to ask users to select options from a menu and validate paths.

✅ Best choice: inquirer
Why? Built-in menus and validation save weeks of work.

// inquirer
const { type } = await inquirer.prompt([
  { type: 'list', name: 'type', choices: ['App', 'Library'] }
]);

Scenario 2: Quick Debug Script

You need to pause a script and grab a variable value instantly.

✅ Best choice: prompt-sync
Why? Minimal code and synchronous flow fits simple scripts.

// prompt-sync
const val = prompt('Enter debug value: ');

Scenario 3: Lightweight CLI Utility

You want a single-file tool with zero dependencies for easy distribution.

✅ Best choice: readline
Why? Core module means no node_modules needed for input.

// readline
const ans = await rl.question('Continue? ');

📌 Summary Table

Feature	`inquirer`	`prompt-sync`	`readline`
Type	External Package	External Package	Core Module
Flow	Async (Promises)	Sync (Blocking)	Async (Promises/Events)
UI Components	Lists, Checkboxes, Input	Text Input Only	Text Input Only
Validation	Built-in	Manual	Manual
Dependencies	Yes	Yes	None

💡 Final Recommendation

Think in terms of user experience and project constraints:

Need rich menus and validation? → Use inquirer. It is the industry standard for professional CLI tools.
Need a quick sync pause in a script? → Use prompt-sync. Keep it limited to simple utilities.
Need zero dependencies? → Use readline. The Promises API makes it viable for modern async code.

These tools each have a place in the Node.js ecosystem. inquirer offers the most features, prompt-sync offers the simplest sync API, and readline offers the most control without external code. Choose based on how much complexity you want to manage yourself.

How to Choose: inquirer vs prompt-sync vs readline

inquirer:
Choose inquirer if you are building professional CLI tools that require rich interactions like selection menus, checkboxes, or input validation. It handles the complexity of terminal UI for you and integrates well with modern async/await code. This is the standard choice for scaffolding tools and installers where user experience matters.
prompt-sync:
Choose prompt-sync for quick scripts or internal tools where blocking the event loop is acceptable and you want minimal setup. It is useful when you need synchronous flow control without managing Promises or event listeners. Avoid this for high-performance servers or complex async applications where blocking I/O is a risk.
readline:
Choose readline when you need maximum control without adding external dependencies to your project. It is ideal for lightweight utilities or when you want to rely solely on Node.js core features. The modern Promises API makes it much easier to use than the older event-based approach while keeping your bundle size down.

Popular Comparisons

inquirer

prompt-sync

readline

README for inquirer

Inquirer.js

A collection of common interactive command line user interfaces.

[!IMPORTANT] This is the legacy version of Inquirer.js. While it still receives maintenance, it is not actively developed. For the new Inquirer, see @inquirer/prompts.

Documentation
1. Installation
2. Examples
3. Methods
4. Objects
5. Question
6. Answers
7. Separator
8. Prompt Types
User Interfaces and Layouts
1. Reactive Interface
Support
Known issues
News
Contributing
License
Plugins

Goal and Philosophy

Inquirer.js strives to be an easily embeddable and beautiful command line interface for Node.js (and perhaps the "CLI Xanadu").

Inquirer.js should ease the process of

providing error feedback
asking questions
parsing input
validating answers
managing hierarchical prompts

Note: Inquirer.js provides the user interface and the inquiry session flow. If you're searching for a full blown command line program utility, then check out commander, vorpal or args.

Documentation

Installation

npm	yarn
`npm install inquirer`	`yarn add inquirer`

import inquirer from 'inquirer';

inquirer
  .prompt([
    /* Pass your questions in here */
  ])
  .then((answers) => {
    // Use user feedback for... whatever!!
  })
  .catch((error) => {
    if (error.isTtyError) {
      // Prompt couldn't be rendered in the current environment
    } else {
      // Something else went wrong
    }
  });

Examples (Run it and see it)

Check out the packages/inquirer/examples/ folder for code and interface examples.

yarn node packages/inquirer/examples/pizza.js
yarn node packages/inquirer/examples/checkbox.js
# etc...

Methods

[!WARNING] Those interfaces are not necessary for modern Javascript, while still maintained, they're depreciated. We highly encourage you to adopt the more ergonomic and modern API with @inquirer/prompts. Both inquirer and @inquirer/prompts are usable at the same time, so you can progressively migrate.

`inquirer.prompt(questions, answers) -> promise`

Launch the prompt interface (inquiry session)

questions (Array) containing Question Object (using the reactive interface, you can also pass a Rx.Observable instance)
answers (object) contains values of already answered questions. Inquirer will avoid asking answers already provided here. Defaults {}.
returns a Promise

`inquirer.registerPrompt(name, prompt)`

name (string) name of the this new prompt. (used for question type)
prompt (object) the prompt object itself (the plugin)

`inquirer.createPromptModule() -> prompt function`

Create a self contained inquirer module. If you don't want to affect other libraries that also rely on inquirer when you overwrite or add new prompt types.

const prompt = inquirer.createPromptModule();

prompt(questions).then(/* ... */);

Objects

Question

A question object is a hash containing question related values:

type: (String) Type of the prompt. Defaults: input - Possible values: input, number, confirm, list, rawlist, expand, checkbox, password, editor
name: (String) The name to use when storing the answer in the answers hash. If the name contains periods, it will define a path in the answers hash.
message: (String|Function) The question to print. If defined as a function, the first parameter will be the current inquirer session answers. Defaults to the value of name (followed by a colon).
default: (String|Number|Boolean|Array|Function) Default value(s) to use if nothing is entered, or a function that returns the default value(s). If defined as a function, the first parameter will be the current inquirer session answers.
choices: (Array|Function) Choices array or a function returning a choices array. If defined as a function, the first parameter will be the current inquirer session answers. Array values can be simple numbers, strings, or objects containing a name (to display in list), a value (to save in the answers hash), and a short (to display after selection) properties. The choices array can also contain a Separator.
validate: (Function) Receive the user input and answers hash. Should return true if the value is valid, and an error message (String) otherwise. If false is returned, a default error message is provided.
filter: (Function) Receive the user input and answers hash. Returns the filtered value to be used inside the program. The value returned will be added to the Answers hash.
transformer: (Function) Receive the user input, answers hash and option flags, and return a transformed value to display to the user. The transformation only impacts what is shown while editing. It does not modify the answers hash.
when: (Function, Boolean) Receive the current user answers hash and should return true or false depending on whether or not this question should be asked. The value can also be a simple boolean.
pageSize: (Number) Change the number of lines that will be rendered when using list, rawList, expand or checkbox.
prefix: (String) Change the default prefix message.
suffix: (String) Change the default suffix message.
askAnswered: (Boolean) Force to prompt the question if the answer already exists.
loop: (Boolean) Enable list looping. Defaults: true
waitUserInput: (Boolean) Flag to enable/disable wait for user input before opening system editor - Defaults: true

default, choices(if defined as functions), validate, filter and when functions can be called asynchronously. Either return a promise or use this.async() to get a callback you'll call with the final value.

{
  /* Preferred way: with promise */
  filter() {
    return new Promise(/* etc... */);
  },

  /* Legacy way: with this.async */
  validate: function (input) {
    // Declare function as asynchronous, and save the done callback
    const done = this.async();

    // Do async stuff
    setTimeout(function() {
      if (typeof input !== 'number') {
        // Pass the return value in the done callback
        done('You need to provide a number');
      } else {
        // Pass the return value in the done callback
        done(null, true);
      }
    }, 3000);
  }
}

Answers

A key/value hash containing the client answers in each prompt.

Key The name property of the question object
Value (Depends on the prompt)
- confirm: (Boolean)
- input : User input (filtered if filter is defined) (String)
- number: User input (filtered if filter is defined) (Number)
- rawlist, list : Selected choice value (or name if no value specified) (String)

Separator

A separator can be added to any choices array:

// In the question object
choices: [ "Choice A", new inquirer.Separator(), "choice B" ]

// Which'll be displayed this way
[?] What do you want to do?
 > Order a pizza
   Make a reservation
   --------
   Ask opening hours
   Talk to the receptionist

The constructor takes a facultative String value that'll be use as the separator. If omitted, the separator will be --------.

Separator instances have a property type equal to separator. This should allow tools façading Inquirer interface from detecting separator types in lists.

Prompt types

Note:: allowed options written inside square brackets ([]) are optional. Others are required.

List - `{type: 'list'}`

Take type, name, message, choices[, default, filter, loop] properties. (Note: default must be set to the index or value of one of the entries in choices)

List prompt

Raw List - `{type: 'rawlist'}`

Take type, name, message, choices[, default, filter, loop] properties. (Note: default must be set to the index of one of the entries in choices)

Raw list prompt

Expand - `{type: 'expand'}`

Take type, name, message, choices[, default] properties. Note: default must be the index of the desired default selection of the array. If default key not provided, then help will be used as default choice

Note that the choices object will take an extra parameter called key for the expand prompt. This parameter must be a single (lowercased) character. The h option is added by the prompt and shouldn't be defined by the user.

See examples/expand.js for a running example.

Checkbox - `{type: 'checkbox'}`

Take type, name, message, choices[, filter, validate, default, loop] properties. default is expected to be an Array of the checked choices value.

Choices marked as {checked: true} will be checked by default.

Choices whose property disabled is truthy will be unselectable. If disabled is a string, then the string will be outputted next to the disabled choice, otherwise it'll default to "Disabled". The disabled property can also be a synchronous function receiving the current answers as argument and returning a boolean or a string.

Checkbox prompt

Confirm - `{type: 'confirm'}`

Take type, name, message, [default, transformer] properties. default is expected to be a boolean if used.

Confirm prompt

Input - `{type: 'input'}`

Take type, name, message[, default, filter, validate, transformer] properties.

Input prompt

Input - `{type: 'number'}`

Take type, name, message[, default, filter, validate, transformer] properties.

Password - `{type: 'password'}`

Take type, name, message, mask,[, default, filter, validate] properties.

Password prompt

Note that mask is required to hide the actual user input.

Editor - `{type: 'editor'}`

Take type, name, message[, default, filter, validate, postfix, waitUserInput] properties

Launches an instance of the users preferred editor on a temporary file. Once the user exits their editor, the contents of the temporary file are read in as the result. The editor to use is determined by reading the $VISUAL or $EDITOR environment variables. If neither of those are present, notepad (on Windows) or vim (Linux or Mac) is used.

The postfix property is useful if you want to provide an extension.

Use in Non-Interactive Environments

prompt() requires that it is run in an interactive environment. (I.e. One where process.stdin.isTTY is true). If prompt() is invoked outside of such an environment, then prompt() will return a rejected promise with an error. For convenience, the error will have a isTtyError property to programmatically indicate the cause.

Reactive interface

Internally, Inquirer uses the JS reactive extension to handle events and async flows.

This mean you can take advantage of this feature to provide more advanced flows. For example, you can dynamically add questions to be asked:

const prompts = new Rx.Subject();
inquirer.prompt(prompts);

// At some point in the future, push new questions
prompts.next({
  /* question... */
});
prompts.next({
  /* question... */
});

// When you're done
prompts.complete();

And using the return value process property, you can access more fine grained callbacks:

inquirer.prompt(prompts).ui.process.subscribe(onEachAnswer, onError, onComplete);

Support (OS Terminals)

You should expect mostly good support for the CLI below. This does not mean we won't look at issues found on other command line - feel free to report any!

Mac OS:
- Terminal.app
- iTerm
Windows (Known issues):
- Windows Terminal
- ConEmu
- cmd.exe
- Powershell
- Cygwin
Linux (Ubuntu, openSUSE, Arch Linux, etc):
- gnome-terminal (Terminal GNOME)
- konsole

Known issues

nodemon - Makes the arrow keys print gibrish on list prompts. Workaround: Add { stdin : false } in the configuration file or pass --no-stdin in the CLI. Please refer to this issue
grunt-exec - Calling a node script that uses Inquirer from grunt-exec can cause the program to crash. To fix this, add to your grunt-exec config stdio: 'inherit'. Please refer to this issue
Windows network streams - Running Inquirer together with network streams in Windows platform inside some terminals can result in process hang. Workaround: run inside another terminal. Please refer to this issue

News on the march (Release notes)

Please refer to the GitHub releases section for the changelog

Contributing

Unit test Please add a unit test for every new feature or bug fix. yarn test to run the test suite.

Documentation Add documentation for every API change. Feel free to send typo fixes and better docs!

We're looking to offer good support for multiple prompts and environments. If you want to help, we'd like to keep a list of testers for each terminal/OS so we can contact you and get feedback before release. Let us know if you want to be added to the list (just tweet to @vaxilart) or just add your name to the wiki

License

Plugins

You can build custom prompts, or use open sourced ones. See @inquirer/core documentation for building custom prompts.

You can either call the custom prompts directly (preferred), or you can register them (depreciated):

import customPrompt from '$$$/custom-prompt';

// 1. Preferred solution with new plugins
const answer = await customPrompt({ ...config });

// 2. Depreciated interface (or for old plugins)
inquirer.registerPrompt('custom', customPrompt);
const answers = await inquirer.prompt([
  {
    type: 'custom',
    ...config,
  },
]);

When using Typescript and registerPrompt, you'll also need to define your prompt signature. Since Typescript is static, we cannot infer available plugins from function calls.

import customPrompt from '$$$/custom-prompt';

declare module 'inquirer' {
  interface QuestionMap {
    // 1. Easiest option
    custom: Parameters<typeof customPrompt>[0];

    // 2. Or manually define the prompt config
    custom_alt: { message: string; option: number[] };
  }
}