webcli
par erdinccurebal
Web & Frontend DevelopmentRésumé
Browse the web, read page content, click buttons, fill forms, take screenshots, and get accessibility snapshots.
Description
Ce que fait ce skill
webcli is a headless browser CLI tool that lets agents navigate websites, read page content, interact with elements, and capture screenshots or PDFs. It runs a persistent browser daemon and exposes commands for clicks, form fills, accessibility snapshots, and JavaScript evaluation.
Unlike raw Playwright scripts, webcli exposes browser control as simple shell commands that an AI agent can chain without writing or managing code files.
Quand l'utiliser
- Scraping product prices from an e-commerce site
- Filling and submitting a web form automatically
- Taking screenshots of a dashboard for reporting
- Logging into a web app and extracting account data
- Testing a web UI flow without writing Playwright code
Exemple de workflow
User asks: go to example.com/login, log in with my credentials, and screenshot the dashboard
- 1
Run
webcli go https://example.com/login - 2
Run
webcli snapshotto get accessibility refs for the email and password fields - 3
Run
webcli fill @e1 'user@example.com'andwebcli fill @e2 'password123' - 4
Run
webcli click 'Log in'thenwebcli sleep 1500 - 5
Run
webcli screenshot -o dashboard.pngto capture the result
Screenshot saved to dashboard.png showing the authenticated dashboard page
Prérequis
- npx playwright install chromium (one-time browser binary install)
Installation
clawhub install erdinccurebal/webcliExécute la commande CLI, ou colle l'instruction dans Claude Code, Cursor, Codex ou l'agent de code que tu utilises.