computer-use

computer-use is a skill for Claude Code, Codex from Panchangam18/background-computer-use. It costs 106 tokens per session (1,987 once invoked), scanned A, original, MIT.

Guidance for controlling macOS applications in the background through their visible controls and accessibility information.

In plain words
What is it for?
Clicking, typing, scrolling, reading interface state, and automating apps such as Safari, Chrome, or Apple Music.
Why use it?
It lets the agent inspect and operate an app without taking focus away from the user or moving their cursor.

Skill for Claude CodeCodex

Part of the background-computer-use plugin — 8 skills, 1 MCP server shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/panchangam18/background-computer-use/computer-use
Any agent
npx skills add Panchangam18/background-computer-use --skill computer-use
Clone the repo
git clone --depth 1 https://github.com/Panchangam18/background-computer-use

Made for: Claude Code, Codex.

Or install background-computer-use, the plugin that ships this one along with the rest of its 8 skills, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for computer-use

README.md
[![agentmods](https://agentmods.dev/badge/skills/panchangam18/background-computer-use/computer-use.svg)](https://agentmods.dev/skills/panchangam18/background-computer-use/computer-use)
Your own site
<a href="https://agentmods.dev/skills/panchangam18/background-computer-use/computer-use"><img src="https://agentmods.dev/badge/skills/panchangam18/background-computer-use/computer-use.svg" alt="Measured on agentmods" height="20"></a>
Per session 106 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,987 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00106 $0.01987
Opus 5 $0.00053 $0.00993
Sonnet 5 $0.00021 $0.00397
Haiku 4.5 $0.00011 $0.00199

Measured 3d ago against content hash 5d91441b9036, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

computer-use scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/computer-use/SKILL.md · 174 lines

How it starts

The opening of the file, as written. The whole thing — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Background computer use (macOS)

You have access to a Codex-style computer-use tool surface that drives macOS applications in the background — without activating them, bringing them to the front, or moving the user's cursor. The tools are provided by the background-computer-use MCP server and are named:

  • list_apps
  • get_app_state
  • click
  • drag
  • perform_secondary_action
  • press_key
  • scroll
  • set_value
  • type_text

The only interaction pattern that works

Use this loop every time:

  1. Discover (optional). If you don't know which app to drive, call list_apps. The app argument on every other tool accepts either the display name ("Finder") or the bundle identifier ("com.apple.finder").
  2. Observe. Call get_app_state(app) once at the start of every turn before acting on that app. Its response contains:
    • an indexed accessibility tree of the key window
    • a window screenshot
    • (sometimes) bundled app-specific hints under <app_hints>
  3. Decide. Pick element indexes from the tree wherever possible. Those indexes are valid only until your next get_app_state call — treat them as snapshot-local, never durable.
  4. Act. Invoke click / type_text / press_key / scroll / set_value / drag / perform_secondary_action using the index you just picked (or raw coordinates as a fallback).
  5. Re-observe after any action that likely changed the UI (menu opened, view navigated, sheet presented, async result arrived). Stale indexes are the #1 cause of silent failures.

If you skip step 2, tools that take element_index will fail with "No captured app state for this app."

Accessibility vs. coordinates

Prefer accessibility-based targeting. It is more robust to window moves, scaling, and theme changes.

  • Use click(app, element_index=N) when the target appears in the tree. For left single-clicks on elements that support AXPress, the server calls AXPress directly, which works even when the app is occluded.
  • Use click(app, x=..., y=...) only when the tree does not surface the target (e.g. custom canvases, WebViews without AX, games). Coordinates are screen points in the top-left origin.
  • Use perform_secondary_action(app, element_index, action) when the element advertises an action like "open", "Increment", "Decrement", "Pick", "ScrollToVisible", "Confirm". Most secondary actions are background-safe.
    • Refused by default: AXShowMenu, AXShowAlternateUI, and AXRaise almost always bring the target app to the foreground because they need to present a visible menu or modal. The tool raises ToolError on these unless you pass allow_foreground_activation=True. Only use that flag when you've confirmed with the user that activating the app is acceptable for this turn.

Read the full file on GitHub · 174 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 174 lines · 106 tokens per session scan A 5d91441b9036

Subscribe to this mod's changes

computer-use is a skill published in the GitHub repository Panchangam18/background-computer-use (5 stars, last pushed 4mo ago), licensed MIT. It adds 106 tokens to every session and 1,987 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

axstream

Fast deterministic macOS UI automation via replayable macros. Use BEFORE driving any macOS app's UI by hand (clicking/typing via computer-use or screenshots) — if a macro exists, replay finishes the task in seconds with no per-step reasoning; after doing a UI task manually, save it as a macro so the next run is fast.…

milind-soni/axstream · 96 tokens

open-computer-use

Platform-neutral guidance for using Open Computer Use, the open-source Computer Use MCP server and CLI for macOS, Linux, and Windows. Use when an agent needs to install, verify, troubleshoot, configure, or operate Open Computer Use through its native CLI, stdio MCP server, or direct Computer Use tool calls.

iFurySt/open-codex-computer-use · 68 tokens

macos-harness

Control a whole Mac from one persistent Python session with screenshots, PID-targeted input, an animated virtual pointer, targeted Apple Accessibility, Apple Events, Browser Harness CDP, and filesystem access. Use for native, Electron, browser, dialog, file, or cross-app tasks without moving the physical cursor or…

browser-use/macos-harness · 71 tokens

m07-concurrency

CRITICAL: Use for concurrency/async. Triggers: E0277 Send Sync, cannot be sent between threads, thread, spawn, channel, mpsc, Mutex, RwLock, Atomic, async, await, Future, tokio, deadlock, race condition, 并发, 线程, 异步, 死锁.

moeru-ai/auv · 75 tokens

m13-domain-error

Use when designing domain error handling. Keywords: domain error, error categorization, recovery strategy, retry, fallback, domain error hierarchy, user-facing vs internal errors, error code design, circuit breaker, graceful degradation, resilience, error context, backoff, retry with backoff, error recovery, transient…

moeru-ai/auv · 95 tokens

m14-mental-model

Use when learning Rust concepts. Keywords: mental model, how to think about ownership, understanding borrow checker, visualizing memory layout, analogy, misconception, explaining ownership, why does Rust, help me understand, confused about, learning Rust, explain like I'm, ELI5, intuition for, coming from Java, coming…

moeru-ai/auv · 91 tokens