Automates browser tasks via CDP with clickable element snapshots and screenshots.
npx skills add https://github.com/iBz-04/gloamy --skill agent-browser-ibz-04
Please help me install this Agent Skill. Skill: agent-browser Source: https://github.com/iBz-04/gloamy/tree/main/.agents/skills/agent-browser Command: npx skills add https://github.com/iBz-04/gloamy --skill agent-browser-ibz-04
This Skill removes the manual effort of interacting with complex websites by letting you navigate, click, fill forms, extract information, and capture visual evidence programmatically.
Instruct your agent to open the login page at https://example.com/login, wait for network idle, snapshot interactive elements, then fill the username and password fields and click the submit button.
High-intent search queries and answers about installing and using this skill.
Web interaction and form filling are automated by taking accessibility snapshots to discover interactive elements via refs, then performing deterministic clicks and inputs. The system waits for network idle or URL patterns and re-snapshots after DOM updates to ensure reliability.
Yes, login sessions can be reused for browser automation by restoring authentication state. You can persist sessions using state files, persistent profiles, session names, or an encrypted auth vault, eliminating the need to repeatedly authenticate during automated workflows.
The best way to capture visual evidence during web scraping is by utilizing built-in screenshot and PDF generation features. You can capture full or annotated screenshots and generate PDFs to verify UI testing outcomes and confirm workflow results.
Yes, you need Chrome or Chromium control via CDP to run these browser automation tasks. The Skill relies on the Chrome DevTools Protocol to navigate URLs, discover interactive elements, and execute deterministic interactions like clicks and form submissions.
Browser automation handles dynamic DOM changes by implementing a ref lifecycle that automatically re-snapshots the page after navigation or DOM updates. This ensures that interactive element references remain accurate and deterministic interactions do not fail on dynamic pages.
Yes, you can run parallel sessions and batch command sequences for large scale web scraping and UI testing. This approach reduces overhead for multi-step tasks by executing multiple automated browser workflows concurrently.
x-cmdAutomate Chrome/Chromium browser interactions via CDP for web application testing.
mupengi-botFetch HTML from URLs and extract structured data into JSON or Markdown.
lazyFrogLOLRecover passwords and credential data from disk images or corrupted files.
nanocoaiAutomate web browsing to extract data, fill forms, and capture screenshots.
