Shortest

**Shortest — Natural Language QA Tests via AI

Shortest is a QA testing tool that lets you author tests in plain English and delegates the execution to AI. Built on Playwright with seamless GitHub integration, it enables you to describe test scenarios in natural language while the system handles implementation details.

Key idea: write human-friendly test scenarios (in plain English) and rely on AI to generate and run the underlying Playwright-based tests.

How it works

1. Write tests in plain English, describing user actions and expectations.
2. Shortest processes the descriptions and generates executable tests using Playwright under the hood.
3. Tests are executed with GitHub integration, allowing you to run in CI or locally via your GitHub workflow.
4. The tool supports environment variables for sensitive data (e.g., username and password) and executes tests with the provided credentials.
5. Results are surfaced back to you, with the ability to view and adjust scenarios as needed.

Example workflow snippet:

shortest(
 'Login to the app using email and password',
 { username: process.env.GITHUB_USERNAME, password: process.env.GITHUB_PASSWORD }
)

Ready to get started? Start writing natural language tests today and let AI handle the implementation. View on GitHub

How to Use Shortest

1. Write a plain-English test description, e.g., "Login to the app using email and password".
2. Provide any necessary parameters (e.g., credentials) via environment variables or a config object.
3. Run the test; AI will generate Playwright-based test code to implement the scenario.
4. Execute tests in your preferred environment (locally or via GitHub Actions).
5. Review results, iterate on test descriptions, and re-run as needed.

Safety and Best Practices

• Avoid embedding sensitive credentials directly in test descriptions.
• Use environment variables or secret management for usernames, passwords, and tokens.
• Keep plain-English steps synchronous and unambiguous to improve AI-generated code quality.

Core Features

• Write tests in plain English and have AI generate executable Playwright tests
• Seamless Playwright integration for robust browser automation
• GitHub integration for CI/CD and workflow automation
• Environment-variable support for secure credentials
• Quick iteration: modify natural language steps and re-run tests
• View and manage test results directly from GitHub and local environments