Skip to content

Automate This

Turns a screen recording of a manual task into proposed scripts and automations

Stars
40,703
Licence
MIT
Files
1 file

What it does

When you give the agent a screen recording of yourself doing a repetitive task, it extracts still frames every few seconds and transcribes any spoken narration. It then reconstructs the steps you took, asks you to confirm them, and proposes automation at up to three levels of effort, from a quick one-liner to a scheduled full script, using tools already on your computer.

Great for

  • People who repeat the same multi-step computer task every week and want it scripted
  • Developers or ops staff who want a quick first draft of automation from a recorded workflow
  • Non-programmers who can narrate a process and want a working script proposed for them

Skip it if

  • Tasks that involve sensitive logins or private data in the recording, since frames and audio are sent into the agent's context and may be stored in its logs
  • Highly complex or unstable workflows where a wrong reconstruction would produce unreliable automation, so the output needs careful human review
  • Users who lack ffmpeg and cannot install it, since the skill cannot extract frames without it

Try asking

  • “Analyze the screen recording at ~/Desktop/monday-report.mov and suggest how to automate it”
  • “Here is a video of me exporting CSVs and emailing them, build a script for it”
  • “Watch this recording of my invoice process and propose a quick win first”
Step-by-step guide: how to use Automate ThisInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    automate-this.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip automate-this.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip automate-this.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Record with clear narration explaining why each step happens, since the transcript reveals pain points and decision points
  • Keep recordings under five minutes so frames are sampled at a readable rate and the reconstruction stays accurate
  • Check the reconstructed step list carefully before approving, because proposals depend entirely on it

What's inside

  • SKILL.md

Redistributed unmodified from github/awesome-copilot under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.