Repterm

Repterm is a terminal-first testing framework designed for CLI and TUI applications. Unlike traditional test runners that mock terminal behavior, Repterm runs every test inside a real PTY (pseudo-terminal) environment, giving you accurate results that reflect how your application actually behaves in a terminal.

Packages

The Repterm monorepo ships three packages:

Key Features

• Real PTY testing -- Tests execute in genuine pseudo-terminal sessions, not simulated environments. What you test is what your users experience.
• Recording mode -- Capture terminal sessions as asciinema .cast files using tmux, perfect for documentation, demos, and visual regression.
• Parallel execution -- Run test suites across multiple workers to cut down execution time on large test suites.
• Plugin system -- Extend Repterm with custom methods, context, and lifecycle hooks through a type-safe plugin API.
• Interactive command testing -- Send keystrokes, wait for prompts, and assert on live terminal output with expect and send.
• Multi-terminal support -- Spawn additional terminal panes within a single test, with automatic tmux pane splitting in recording mode.

Quick Example

import { test, expect } from 'repterm';

test('echo command', async ({ $ }) => {
  const result = await $`echo "Hello, Repterm!"`;
  expect(result).toSucceed();
  expect(result).toHaveStdout('Hello, Repterm!');
});

The $ tagged template is the primary API for running commands. It returns a CommandResult with structured access to exit code, stdout, stderr, combined output, and duration.

Next Steps

• Installation -- Set up Repterm and system dependencies
• Getting Started -- Write and run your first test in under 5 minutes
• Architecture -- Understand how the layers fit together
• API Reference -- Full DSL and assertion reference