Name: Peekaboo
Author: steipete

What breaks without openclaw screenshot skill macos

Agents without visual context. Screen state invisible. LLM vision capabilities unused.

→

Agent visual awareness × 165-star macOS skill ÷ 5–10 minute setup ÷ no third-party upload = screen-aware automations.

openclaw screenshot skill macos — what it actually does

Capture the current screen or webcam image from OpenClaw.

Pass screenshots directly to vision-capable LLM agents.

Build visual feedback loops where agents respond to screen state.

Resize images before sending to control LLM call cost.

Requires Screen Recording permission — one-time setup.

Security check — openclaw screenshot skill macos

Privacy score: 7/10 — accesses connected platform APIs only. Lock it: review OAuth scopes before install, confirm macOS 12+; OpenClaw ≥1.0 compatibility.

Quick start — openclaw screenshot skill macos in 5–10 minutes

Setup time: 5–10 minutes

You need:

OpenClaw core
macOS 12+
Screen Recording permission granted

Install the package:

clawhub install steipete/peekaboo

Grant Screen Recording permission in System Preferences

Install skill

Run /peekaboo screenshot to capture current screen

Image is passed to vision-capable LLM automatically

Troubleshooting openclaw screenshot skill macos

1. Screen Recording permission denied silently — check System Preferences > Privacy

2. Large screenshots slow down LLM calls — resize before sending

Compatibility & status

Works with: macOS 12+; OpenClaw ≥1.0 beginner Last updated: Oct 2025 ★ 165 on GitHub MIT

Official docs →

View on GitHub →

Related — more like openclaw screenshot skill macos

More by steipete

A blind agent misses everything happening on screen. Install the openclaw screenshot skill macos before your next visual workflow.

Get it on GitHub →