skills.rest
roseDwayane
roseDwayaneCOMMUNITY
Explore all skills by this author →
GitHub

agent-browser

Automates browser navigation, element interaction, screenshots, and data extraction via command-line commands.

1|Updated May 22, 2026
Search Tags:#data extraction#screenshots#browser automation#web testing#web scraping#form filling#agent-browser
One-click install
npx skills add https://github.com/roseDwayane/LocalizeAgenticSys --skill agent-browser-rosedwayane
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/roseDwayane/LocalizeAgenticSys/tree/main/nanoclaw/container/skills/agent-browser
Command: npx skills add https://github.com/roseDwayane/LocalizeAgenticSys --skill agent-browser-rosedwayane

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manual web browsing for research, form filling, and data collection is repetitive and time-consuming. This Skill lets an AI agent drive a real browser through simple commands, handling navigation, interaction, and extraction without manual effort. ## Core Features & Use Cases - Page Navigation & Interaction: Open URLs, click elements, fill forms, press keys, and scroll using snapshot refs or semantic locators. - Data Extraction & Capture: Read element text, attributes, and page metadata, plus take screenshots and export pages as PDF. - Session Persistence: Save and reload authentication state, cookies, and localStorage to reuse logged-in sessions. - Use Case: Research a topic by opening multiple articles, extracting key text from each page, and saving screenshots as evidence — all through scripted browser commands. ## Quick Start Use the agent-browser skill to open a webpage, snapshot its interactive elements, and extract the information I need.

Frequently Asked Questions about agent-browser

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate browser interactions from the command line?▼

Use agent-browser to open a URL, run snapshot -i to list interactive elements with refs like @e1, then click, fill, or type against those refs. Re-snapshot after navigation or major DOM changes to get updated refs.

How to fill and submit a web form programmatically?▼

Open the form page, snapshot with -i to find input refs, then use fill for each field and click the submit button. Finish with wait --load networkidle and a new snapshot to verify the result.

Can I reuse a logged-in browser session across runs?▼

Yes. After logging in once, run agent-browser state save auth.json to persist the session. Later, load it with state load auth.json before opening authenticated pages, avoiding repeated logins.

How do I wait for dynamic page content to load?▼

Use agent-browser wait with a target: an element ref, a millisecond delay, --text for specific content, --url for URL patterns, or --load networkidle for network quiescence. This handles SPAs and lazy-loaded content.

What are alternatives to element refs for locating page elements?▼

Use semantic locators via the find command: locate by role, text, label, or placeholder, such as find role button click --name "Submit". This is more resilient when refs change between snapshots.

Related Skills

Explore More Software Engineering Skills→
citedycitedy
31

cron

Manage recurring tasks via the adclaw CLI using cron jobs.

Official
Advanced
flaz78flaz78
9

pdf_reader

Extract readable text from selectable-content PDF documents.

Community
Intermediate
MilUX-LtdMilUX-Ltd

sector-scan

Refresh sector reference pages by re-running web research and reporting changes.

Official
Intermediate
richardnguyen0715richardnguyen0715

scrapling

Scrape static and JavaScript-rendered web content with stealth and Cloudflare bypass.

Community
Advanced