Menu
Automated visual testing with Playwright MCP - test web apps, presentations, websites, and documents with scalable reviewer perspectives
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Expert guidance on RAG (Retrieval-Augmented Generation) system design including chunking strategies, embedding selection, retrieval methods, and vector database choices
Design complex system architectures, evaluate tradeoffs, and make critical technical decisions requiring deep reasoning
Deep code analysis identifying subtle bugs, security issues, performance problems, and architectural concerns requiring expert-level reasoning
Efficient context state inspection, task lifecycle management, and session tracking
Systematic codebase onboarding. Builds a mental model of a new or unfamiliar project by exploring structure, architecture, key data models, entry points, and auth patterns.
Systematically extract knowledge from scattered documentation to prevent catastrophic forgetting. Creates structured extraction reports with status tags.
| name | gui-test |
| description | Automated visual testing with Playwright MCP - test web apps, presentations, websites, and documents with scalable reviewer perspectives |
| short_desc | Playwright MCP visual + screenshot testing |
| keywords | ["visual test","Playwright MCP","screenshot diff","web app test","UI test","Playwright","visual regression","automated GUI test","test web app","screenshot comparison","visual diff","presentation test","document test"] |
| model | sonnet |
Runs visual testing using the gui_testing framework + Playwright MCP. Supports web apps, presentations, websites, and documents with configurable complexity.
quick — aesthetics/layout only (1 agent, fastest)standard — aesthetics + usability + bug detection (3 agents)full — all 5 perspectives (default for web apps)web_app — default, UI testing (Tauri, Gradio, etc.)presentation — slide design, readability, visual hierarchy, brand consistencywebsite — layout, navigation, responsiveness, content claritydocument — formatting, typography, readability, structure/gui-test # full suite, web_app (localhost:1420)
/gui-test http://localhost:7861 # full suite, specific URL
/gui-test http://localhost:1420 CommsTab # full suite, focus on component
/gui-test presentation slides.html quick # quick presentation check
/gui-test website https://vibecodedtools.it standard # standard website review
/gui-test document http://localhost:8080/report quick # quick document check
Parse arguments:
presentation, website, document), use it; next token is URLquick, standard, full) appears, use itRun via the MAO gui_testing framework:
# POSIX (bash/zsh)
python -m orchestrator.gui_testing.runner \
--url {URL} \
--complexity {quick|standard|full} \
--target-type {web_app|presentation|website|document} \
--output-dir "$(python -c 'import tempfile,os;print(os.path.join(tempfile.gettempdir(),"vct-gui-report"))')"
# Windows (PowerShell)
python -m orchestrator.gui_testing.runner `
--url {URL} `
--complexity {quick|standard|full} `
--target-type {web_app|presentation|website|document} `
--output-dir "$env:TEMP\vct-gui-report"
The output dir is the vct-gui-report folder under the system temp dir — same name the gui-tester agent uses for its screenshots, so reports and screenshots land together.
After the agent completes, read the report and show the user: