An OpenClaw workflow skill that enables AI agents to coordinate screen captures, OCR, screen recordings, and annotations using the CleanShot Tool plugin.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install cleanshot-skill
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install cleanshot-skill using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is CleanShot Skill for OpenClaw?
The CleanShot Skill is an optional workflow and decision-making layer designed for OpenClaw agents. Rather than executing terminal operations directly, this skill guides your AI assistant on choosing precise tool configurations and parameters provided by the openclaw-plugin-cleanshot package. It bridges the gap between natural language requests and native macOS screen utility operations, allowing for a seamless visual automation workflow.
By adding this to your workspace of Openclaw Skills, your AI companion gains advanced visual spatial reasoning. It can automatically calculate multi-monitor coordinates, handle complex capture modes (such as scrolling captures and previous-area crops), run OCR text extractions, and initiate on-screen markup processes dynamically, making it an indispensable asset for developers, QA testers, and power users alike.
CleanShot Skill for OpenClaw Use Cases
- Triggering swift captures of the active screen or window directly from a chat interface and copying it to the clipboard.
- OCR extraction of text from specific regions, automatically parsing line breaks for structured copying.
- Multi-monitor screenshot capturing where the AI intelligently calculates left/right halves or specific thirds of external displays.
- Creating animated GIFs or MP4 recordings of live UI interactions for bug tracking and documentation.
- Automating scrolling captures of full web pages, lengthy terminal outputs, or long nested chat histories.
How CleanShot Skill for OpenClaw Works
- User Prompt Parsing: The user requests a visual action, such as 'take a screenshot of the second monitor's left half' or 'extract this text'.
- Workflow Evaluation: The AI evaluates the instruction against defined decision rules within the skill, selecting the correct tool from the active CleanShot plugin.
- Coordinate Calculation: If multi-monitor parameters or partial regions are specified, the agent calls the display geometry tool and computes precise x, y, width, and height dimensions.
- Plugin Execution: The agent dispatches the parameterized API call to openclaw-plugin-cleanshot (e.g., cleanshot_capture with custom coordinates and an action like copy, save, or annotate).
- State Feedback: The screen capture tool triggers native CleanShot X operations and feeds the success status back to the agent interface.
CleanShot Skill for OpenClaw Setup
To leverage this workflow behavior, ensure the CleanShot Tool plugin is installed in your runtime. It is a vital part of the broader catalog of Openclaw Skills designed for macOS productivity.
Install the required plugin:
npm install openclaw-plugin-cleanshot
Ensure that CleanShot X is configured to allow external applications to control its API. Navigate to:
CleanShot X -> Settings -> Advanced -> API -> Allow Applications to control CleanShot
CleanShot Skill for OpenClaw Data Schema & Taxonomy
The CleanShot Skill organizes capture modes, destinations, and coordinate models as follows:
| Tool Parameter | Expected Types / Values | Purpose |
|---|---|---|
mode |
fullscreen, window, area, previous_area, self_timer |
Directs the type of screen capture interaction |
action |
copy, save, annotate, pin, upload |
Determines output routing of the visual asset |
coordinates |
x, y, width, height (integers) |
Precise geometry of targeted display regions |
display |
Integer (e.g., 1, 2) |
Targeted display identifier for multi-monitor setups |
Spatial Coordinate Logic
Calculations map logical screens into bounding boxes dynamically:
- Left Half:
x = d.x,y = d.y,width = d.width / 2,height = d.height - Right Half:
x = d.x + d.width / 2,y = d.y,width = d.width / 2,height = d.height - Thirds: Dynamically calculates standard column positions based on
d.width / 3offsets.
CleanShot Skill for OpenClaw Advanced Features
- Multi-Monitor Coordinate Engine: Auto-discovers active display setups, geometry parameters, and calculates precise fractions of target displays.
- Sequential Screen Monitoring: Employs
previous_areamechanics to capture identical UI coordinates sequentially without repeated manual prompting. - Direct OCR Parsing: Instructs the agent to automatically utilize
linebreaks: trueparsing for fast clipboard text ingestion. - Adaptive Image Fallbacks: Safely shifts from inline chat attachments to secure local directories or clipboard copying if the host chat environment lacks direct media attachment capabilities.
SKILL.md
Loading
Related Openclaw Skills
Dev Guardrails
A universal behavior and guardrail layer for AI coding agents that enforces engineering discipline, prevents scope creep, and guarantees production-viable code quality.
The Creative Act
An executable toolkit modeled on Rick Rubin's creative philosophy to help creators overcome blocks, tune into inspiration, and cultivate a creative mindset.
The Creative Habit
An executable AI-driven toolkit based on Twyla Tharp's philosophy to transform creativity from a mystical whim into a repeatable daily discipline.
Snaplii Autopilot
Snaplii Autopilot is a comprehensive agent-to-merchant workflow that automates buying gift cards with cashback and completes checkouts autonomously via browser control.
Skill Optimizer
An opinionated tool designed to audit, polish, and optimize existing AI agent SKILL.md files against the agentskills.io standard.
The Coddling of the American Mind
An AI-powered cognitive toolkit that implements the behavioral science and psychological principles from Greg Lukianoff and Jonathan Haidt's book to foster resilience, combat cognitive distortions, and build antifragility.
Featured*
