An automated mobile and H5 UI auditor that analyzes layouts for low light, glare, glove usage, and screen moisture, outputting execution-ready handoff documentation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install low-visibility-fix
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 low-visibility-fix 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 Low-Visibility UI Auditor?
The low-visibility-fix agent skill is an engineering-focused auditor optimized for physical and environmental constraints. Designed specifically to analyze existing mobile web pages, WeChat mini-programs, and web applications, it checks for design limitations that affect real-world field operations. Whether workers are operating in extreme glare, dark environments, or handling screens with gloves or wet hands, this tool evaluates tap-targets, contrast ratios, and color-only states against strict field conditions.
As a core utility among Openclaw Skills, this agent focuses purely on diagnostics and reporting. It strictly isolates its actions to analyze code and generate structured documents without modifying any target code. This separation of duties makes it safe to run across complex, critical front-ends, feeding structured recommendations directly to developers or code-modification agents.
Low-Visibility UI Auditor Use Cases
- Evaluating outdoor and industrial field-worker interfaces for glare, high-vibration, and glove-friendly touch targets.
- Checking WeChat mini-programs (WXML/WXSS) and H5 layouts for compliance against low-light contrast and size thresholds.
- Generating non-intrusive, ready-to-implement design audit reports for QA review cycles.
- Creating a structured, machine-readable design handoff to pipe into downstream automated coding agents.
How Low-Visibility UI Auditor Works
- Scope Verification: Validate that the target is an existing mobile UI and assess parameters like pages or CSS selectors to establish audit bounds.
- Preflight Check: Locate critical files such as HTML, WXML, CSS, and WXSS, selecting the optimal visual-pass rendering module.
- Deterministic Audit: Run static scans comparing elements to normalized standards specified in the design-tokens reference file.
- Visual Verification Pass: Render target pages under simulated mobile viewports to evaluate contrast and resolve any complex spatial layouts.
- Handoff Synthesis: Compile the final output as a standardized audit.json and report.md schema inside the output directory.
- Handoff Report: Deliver a clear, prioritized summary categorized by severity levels, ready for human engineers or alternative implementation agents.
Low-Visibility UI Auditor Setup
System Requirements
Ensure Python 3.x is configured in your execution environment.
Running the Audit
Execute the analyzer by passing the directory of the target mobile web or mini-program files:
# Run a comprehensive visual and static analysis on targeted pages
python3 scripts/audit.py <target_directory> --pages homepage,settings --out ./.lv-audit/
Performing Static Analysis on Single Files
Evaluate individual components or layout pages deterministically:
python3 scripts/analyze.py <file_path> --viewport-px 375
Low-Visibility UI Auditor Data Schema & Taxonomy
The skill generates structured, machine-parseable, and human-readable handoff documentation strictly isolated in the output directory (default <target>/.lv-audit/).
Core Handoff Artifacts
| Artifact | Purpose | Output Format |
|---|---|---|
audit.json |
Comprehensive structured dataset of findings segmented by design-token failures. | JSON schema |
report.md |
Prioritized, severity-mapped report utilizing compliance-ready snippet references. | Markdown |
design-tokens.json |
Single source of truth for contrast, sizing, padding, and state thresholds. | Configuration JSON |
Low-Visibility UI Auditor Advanced Features
- Environment-Specific Customization: Matches UI checks directly against realistic environmental constraints such as wet hands, glare, and glove usage using the field-conditions registry.
- Zero-Modification Guarantee: Operates entirely with read-only target access, eliminating direct-write risks to production repositories.
- Graceful Visual Degradation: Falls back to rigorous static code analysis when visual browser drivers or render engines are missing.
- Pipeline Readiness: Interoperates naturally with other Openclaw Skills, generating machine-readable schemas to drive downstream code remediation steps.
SKILL.md
Loading
Related Openclaw Skills
Telegram Crawler & Monitoring Agent (tg-crawler)
An advanced, multi-account Telegram monitoring and intelligence gathering tool built with Telethon that features SOCKS5 proxies, rate-limit recovery, and automated SQLite retention policies.
Image Upload to img.scdn.io
A command-line skill to instantly upload local images to img.scdn.io and retrieve a permanent public URL for sharing.
HK Weather Info
A zero-dependency weather skill leveraging the Hong Kong Observatory OpenData API to provide live weather reports, forecasts, and regional data across multiple languages.
NetLoan Aid
An AI-powered local loan analyzer and repayment assistant designed to evaluate loan compliance, structure repayment strategies, and generate platform negotiation scripts.
Fact-Check Skill
A high-speed, source-backed verification skill designed to triage, execute parallel web queries, and deliver bottom-line answers within a strict time budget.
Tiane Cooking Order Skill
An AI agent skill that automates the end-to-end booking of professional on-demand home cooking and hourly chef services from Swan Home.
Featured*
