What does Alumnium do?

Alumnium enhances test automation with AI by translating natural language instructions into actions for web and mobile apps, working on top of tools like Appium, Playwright, and Selenium.

Who is Alumnium best for?

Alumnium is best for software and test engineers who want to speed up browser and mobile test creation while keeping control over test logic.

Do I need to install anything to use Alumnium?

Yes, you need to install Alumnium as a Python library and wrap it around your existing Appium, Playwright, or Selenium instance.

Does Alumnium offer an API or integrations?

Yes, Alumnium offers integrations with Appium, Playwright, and Selenium, and supports the Model Context Protocol for use with general-purpose AI agents.

How do I contact Alumnium support?

Contact Alumnium support through their documentation at https://alumnium.ai/docs/ or check the website for additional channels.

Newsletters.ai — Learn about AI, the lazy way.

Alumnium

Software Development

An open-source Python library that converts human-readable test instructions into Appium, Playwright, and Selenium actions for web and mobile QA teams.

Available Via:APIWeb

Visit Alumnium

Alumnium is an open-source Python library for software and test engineers who want to write human-readable test instructions and execute them as automated web and mobile tests.

Key capabilities include:

Natural-language test commands using simple primitives like do (actions), check (assertions), and get (data extraction) that translate into executable automation.
Tooling integrations with Selenium, Playwright, and Appium, enabling browser and mobile testing without switching ecosystems.
LLM provider support across Anthropic, Google Gemini, OpenAI, Meta Llama, DeepSeek, and Mistral, plus local models via Ollama.
Reliability features such as automatic waiting and retry behavior during execution.

Pricing: Free (open-source), distributed as a PyPI package (requires Python 3+).

Notable for its high-level abstraction that supports gradual adoption in existing Python test setups, plus an optional Model Context Protocol (MCP) server that lets general-purpose AI agents (for example, Claude Code) drive web and mobile automation. It also reports 98.5% on the WebVoyager benchmark with Claude Code and Selenium.

TransparencySee listing history, review integrity, and health signals

Feature:AI Agents Code Generation Debugging Workflow Automation

Use Case:Writing Code Docs

Audience:Developers

Pricing:Free Open Source