Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

AUV Computer Control

Automates native desktop apps through the moeru-ai/AUV CLI using a strict observe → act → verify loop.

AutomationAdvanced★ 430⑂ 46AI score 7/10Last updated: Oct 7, 2026

What it does

  • Uses the auv CLI (or an already-connected AUV MCP) as the execution layer for native desktop app control.
  • Forces capability discovery first: auv doctor, auv plugin list, auv invoke <command-id> --help — installed help is authoritative, no invented commands.
  • Grounds the target window via window.list --json, capture and OCR, sends the smallest targeted input, then re-observes the semantic effect rather than trusting the return code.
  • Parses the --json envelope (run_id, status, result, artifacts, failure) and keeps stdout/stderr even on nonzero exit.
  • Handles classic failure modes: logical points vs capture pixels, lost focus, popups outside an app-only capture, denied OS permissions, and unsafe retries of send/submit/purchase actions.

Who it's for

  • Desktop automation engineers already using or evaluating AUV.
  • Anyone automating native apps (chat clients, editors, internal terminals) that browser tooling cannot reach.
  • QA/RPA practitioners who must distinguish "input delivered" from "goal achieved".

Example uses

  1. Send a chat message: list windows, disambiguate the conversation, focus the text field, type, verify field contents before Return, then capture the sent bubble as evidence.
  2. Pick a native search result: when an app-only capture shows no results, inspect the full display for a separate popup window, map coordinates with the returned scale/bounds, click, then confirm by window title.
  3. Reuse a proven workflow: turn a verified flow into a minimal task-scoped CLI script with argument arrays, bounded waits and semantic checks instead of blind click/sleep chains.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/aiskillstore/marketplace/HEAD/pending/0xselenicdove/auv-computer-control/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the pending/0xselenicdove/auv-computer-control folder from the GitHub repo aiskillstore/marketplace into my ~/.claude/skills/auv-computer-control/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/aiskillstore/marketplace.git && mkdir -p ~/.claude/skills/auv-computer-control && cp -r marketplace/pending/0xselenicdove/auv-computer-control/* ~/.claude/skills/auv-computer-control/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the repo: git clone https://github.com/aiskillstore/marketplace.git
  3. Create the skill folder: mkdir -p ~/.claude/skills/auv-computer-control
  4. Copy the files: cp -r marketplace/pending/0xselenicdove/auv-computer-control/* ~/.claude/skills/auv-computer-control/
  5. Confirm references/operations.md came along — it holds the AUV setup notes.
  6. Install the moeru-ai/AUV CLI separately; installing the skill does not install AUV. Verify with auv --version and auv doctor.
  7. On macOS, grant Screen Recording and Accessibility permissions to your terminal/app in System Settings → Privacy & Security.
  8. Restart Claude Code and ask something like "use AUV to do X in app Y".
View source on GitHub ↗