← Back to Skills Marketplace
1342
Downloads
0
Stars
4
Active Installs
1
Versions
Install in OpenClaw
/install mac-node-snapshot
Description
A robust, permission-friendly method to capture macOS screens via OpenClaw screen.record. Ideal for headless environments or ensuring capture reliability.
README (SKILL.md)
mac-node-snapshot
Overview
Uses node screen.record to record a 1-second clip and extract a high-quality PNG frame. This workflow bypasses common screencapture permission issues and ensures a reliable image return.
Quick start (single command, no scripts)
All paths are relative to {skill}.
mkdir -p "{skill}/tmp" \
&& openclaw nodes screen record --node "\x3Cnode>" --duration 1000 --fps 10 --no-audio --out "{skill}/tmp/snap.mp4" \
&& ffmpeg -hide_banner -loglevel error -y -ss 00:00:00 -i "{skill}/tmp/snap.mp4" -frames:v 1 "{skill}/tmp/snap.png"
When to use (trigger phrases)
Use this skill when the user asks:
- "Take a screenshot"
- "What is on my screen?"
- "Capture the screen"
- "Screenshot via screen.record"
Notes
- Requirements:
ffmpeg(ask before installing). - If the frame is black, ask the user to wake the screen and retry.
- Use
readon{skill}/tmp/snap.pngto attach it to the reply.
Troubleshooting
- screen_record fails (node disconnected): check
nodes status, ensure OpenClaw app is running/paired. - screenRecording false: must grant Screen Recording in System Settings; cannot be bypassed.
- Black frame: screen may be asleep/locked; ask the user to wake and retry.
Usage Guidance
This instruction-only skill appears coherent and limited to taking a short screen recording and extracting a frame. Before installing or using it: 1) confirm you have the OpenClaw CLI available and ffmpeg installed (the SKILL.md requires both even though the registry metadata omitted them); 2) verify you trust the skill source — provenance is unknown and there is no homepage; 3) be aware this will attempt to capture your screen when invoked, and macOS Screen Recording permission is required (the skill cannot bypass this); 4) if you enable autonomous invocation for agents, they could run the capture when triggered by the listed phrases — only enable that if you trust the agent's prompts. If you want stronger assurance, ask the publisher to update metadata to declare required binaries and provide a homepage or source repository.
Capability Analysis
Type: OpenClaw Skill
Name: mac-node-snapshot
Version: 1.0.0
The skill `mac-node-snapshot` is designed to capture a screenshot on macOS using the `openclaw nodes screen record` command and `ffmpeg`. All operations are confined to the skill's temporary directory (`{skill}/tmp`), and the `SKILL.md` instructions clearly guide the AI agent to perform this specific task, handle errors, and return the image. There is no evidence of data exfiltration, malicious execution, persistence, or prompt injection attempting to subvert the agent's behavior beyond its stated purpose. The skill explicitly acknowledges the need for system permissions for screen recording.
Capability Assessment
Purpose & Capability
The SKILL.md explicitly uses the OpenClaw CLI (openclaw nodes screen record) and ffmpeg to produce a PNG from a short recording, which matches the name/description. However, the registry metadata lists no required binaries while the instructions clearly require the OpenClaw CLI and ffmpeg — a metadata omission that should be corrected before trusting automatic installs or runs.
Instruction Scope
Instructions are narrow and concrete: create a tmp folder under the skill, run openclaw nodes screen record, convert with ffmpeg, and read the output file to attach. They do not ask the agent to read unrelated files, environment variables, or transmit data to third-party endpoints. The SKILL.md also explicitly notes that Screen Recording permission cannot be bypassed.
Install Mechanism
There is no install spec (instruction-only), so nothing is written to disk by an installer. This is lower risk than arbitrary downloads. The only runtime dependencies are external binaries (openclaw CLI and ffmpeg) which the skill expects to be present or be installed with user consent.
Credentials
The skill does not request environment variables, credentials, or config paths. That is proportionate for a screen-capture helper. No unexplained secret access is requested.
Persistence & Privilege
The skill does not request 'always: true' and is not claiming elevated persistent privileges. Agent autonomous invocation is allowed by default (disable-model-invocation: false), which is normal; nothing in the skill attempts to modify other skills or system-wide settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install mac-node-snapshot - After installation, invoke the skill by name or use
/mac-node-snapshot - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release
Metadata
Frequently Asked Questions
What is mac-node-snapshot?
A robust, permission-friendly method to capture macOS screens via OpenClaw screen.record. Ideal for headless environments or ensuring capture reliability. It is an AI Agent Skill for Claude Code / OpenClaw, with 1342 downloads so far.
How do I install mac-node-snapshot?
Run "/install mac-node-snapshot" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is mac-node-snapshot free?
Yes, mac-node-snapshot is completely free (open-source). You can download, install and use it at no cost.
Which platforms does mac-node-snapshot support?
mac-node-snapshot is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created mac-node-snapshot?
It is built and maintained by taozhe6 (@taozhe6); the current version is v1.0.0.
More Skills