2 min read

Claude can record screen actions as reusable skills

Claude can now record screen actions, text input, and voice guidance as reusable skills for repetitive workflows.

Image: kod

Anthropic has added Skill Recording to Claude, turning ordinary on-screen actions into a reusable workflow. The chatbot records clicks, text entry, and voice instructions, then reproduces them as a ready-made routine.

The feature is already available in Claude’s desktop app for Pro, Max, and Team subscribers. Users show Claude a process once, explain it aloud, and save the sequence as a skill—removing the need to write long instructions for repetitive tasks. Android Authority first reported the update.

Menu for adding skill files in ChatGPT and Claude, Android Authority
Menu for adding skill files in ChatGPT and Claude, Android Authority

Recording starts in the Cowork section: open a chat, press the plus button, and select the relevant option. Claude then captures both interactions and spoken explanations, assembling them into an algorithm that can be reused for tasks such as report collection and other fixed-step routines.

Recommended reading

OpenAI models hacked Hugging Face to cheat a benchmark

The feature currently works only on Pro, Max, and Team plans. Anthropic has not said whether it will come to free accounts. Claude joins other services moving beyond chat toward interface-based actions, but its approach focuses on learning from a screen recording rather than manually assembling a workflow.

Sergey Kuznetsov is editor-in-chief of itzine.ru and a technical journalist with 15 years of experience. He specializes in in-depth testing of audio equipment, camera gear, and consumer electronics. He has written more than 5,000 articles covering topics from gaming chairs and networking equipment to analysis of the smartphone and wearable-device markets. At the publication, he oversees expert sections and sets editorial policy.

Ava Chen

AI Editor

Ava covers the rapidly evolving world of artificial intelligence, from foundational models and research labs to the real-world economics of intelligence. With a background in computational linguistics, she cuts through the hype to find out what actually works. She firmly believes that benchmarks are just marketing until reproduced in the wild.

via ITzine

/ Keep reading