peekaboo

Capture and inspect macOS screens to target UI elements for automated interaction.

Updated Apr 5, 2026
One-click install
npx skills add https://github.com/dingdyan/openclaw-workspace-v2 --skill peekaboo-dingdyan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: peekaboo
Source: https://github.com/dingdyan/openclaw-workspace-v2/tree/main/skills/peekaboo
Command: npx skills add https://github.com/dingdyan/openclaw-workspace-v2 --skill peekaboo-dingdyan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Peekaboo removes the need for fragile, manual UI clicking by letting you capture the screen and target the right UI elements reliably for automated interaction on macOS.

Core Features & Use Cases

  • Screen capture & inspection: Capture screens or regions, annotate results, and use the snapshot cache to target elements deterministically.
  • UI element interaction: Click, drag, scroll, swipe, type, press keys, and trigger hotkeys against identified UI elements.
  • App/window/menus control: Launch, manage windows and Spaces, and navigate menus, Dock, and menubar items for end-to-end workflows.
  • Vision mapping for targeting: Use visual maps from snapshots to find elements like buttons, fields, or menus before you act.

Quick Start

Run peekaboo permissions, then capture and inspect the target screen with peekaboo see --app Safari --window-title "Login" --annotate --path /tmp/see.png so you can click the identified element and type credentials safely.

Frequently Asked Questions about peekaboo

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate macOS UI interactions without fragile manual clicking?

macOS UI automation captures and inspects the screen, then targets specific UI elements for reliable input control. You capture a snapshot, annotate it to identify buttons or fields, and execute deterministic click, type, and keypress actions.

What's the best way to script repetitive login automation on macOS?

Login automation on macOS is scripted by capturing the target window, identifying credential fields from the snapshot, and executing targeted input actions. You can save these workflows as .peekaboo.json scripts for repeated deterministic execution.

Can I capture and inspect specific macOS app windows for element targeting?

Yes, screen capture can target specific apps and windows using flags like --app and --window-title. You annotate the captured snapshot to visually map UI elements such as buttons and fields before executing interaction commands.

Does this macOS UI automation tool support keyboard and mouse control?

Keyboard and mouse control is fully supported, allowing you to click, drag, scroll, swipe, type, and trigger hotkeys against identified UI elements. This enables comprehensive end-to-end workflows across apps, windows, and menubar items.

How do I get structured JSON output from a macOS screen capture script?

Structured JSON output is generated by running the screen capture and inspection process with the --json runtime flag. This provides deterministic element targeting data and snapshot analysis results in a machine-readable format.

What permissions do I need to set up macOS UI automation and screen capture?

macOS UI automation requires running a permissions setup command before capturing screens or controlling interfaces. This grants the necessary accessibility and screen recording access for reliable element targeting and input control.