skills.rest
dmuhoro
dmuhoroCOMMUNITY
Explore all skills by this author →
GitHub

browser

Automates web browser navigation, interaction, and data extraction using AI-optimized accessibility snapshots.

Updated Mar 25, 2026
Search Tags:#browser automation#web scraping#form filling#sessions#agent-browser#mcp tools#accessibility snapshot
One-click install
npx skills add https://github.com/dmuhoro/EasyTutor --skill browser-dmuhoro
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: browser
Source: https://github.com/dmuhoro/EasyTutor/tree/main/archive/.claude/skills/browser
Command: npx skills add https://github.com/dmuhoro/EasyTutor --skill browser-dmuhoro

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Automating browser tasks with raw DOM dumps overwhelms AI context windows and makes element targeting unreliable. This Skill reduces context usage by 93% by using accessibility-tree snapshots with stable element refs (@e1, @e2) instead of full HTML. ## Core Features & Use Cases - AI-Optimized Snapshots: Capture interactive elements only (snapshot -i) with deterministic element refs for reliable clicking and form filling. - Full Interaction Set: Navigate, click, fill, type, select, scroll, wait for elements/text/URLs, and capture screenshots. - Isolated Sessions: Run parallel browser sessions with saved authentication state for swarm-style multi-agent workflows. - Use Case: Automate a login flow — open the page, snapshot interactive elements, fill credentials by ref, click submit, and wait for the dashboard URL. ## Quick Start Ask the agent to open a website with the browser skill, take an interactive snapshot, and click or fill elements using the returned refs.

Frequently Asked Questions about browser

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

FAQPage Schema
How do I automate browser interactions with an AI agent?▼

Open the target URL with agent-browser, run snapshot -i to get interactive elements with refs like @e2, then use click, fill, or type commands against those refs. Re-snapshot after any page change to get updated refs.

How to fill out and submit a web form programmatically?▼

Navigate to the form page, take an interactive snapshot, then use fill with each input's element ref and value. Click the submit button's ref and use wait --text or wait --url to confirm submission succeeded.

What is the difference between element refs and CSS selectors in browser automation?▼

Element refs (@e1, @e2) come from accessibility-tree snapshots and are deterministic for the current page state, making them more reliable for AI agents. CSS selectors and semantic locators (role, label, testid) are also supported when refs are unavailable.

Can I run multiple browser sessions in parallel?▼

Yes, use the --session flag to create isolated sessions, each with its own browser state. You can save authentication state from one session and load it into another for coordinated multi-agent scraping.

Why does my browser automation fail after navigation?▼

Element refs become stale after page navigation or DOM changes. Always re-run snapshot after navigation, and use wait commands for elements, text, URLs, or network idle states before interacting.

Related Skills

Explore More Software Engineering Skills→
rayzrurayzru

auth-verify

Authenticate to a web application and verify session state via Chrome DevTools session sharing.

Community
Advanced
ruvnetruvnet
70.1k

cost-report

Generates token usage and USD cost reports aggregated by agent, model, and tier.

Community
Intermediate
CodingHeaderCodingHeader

verification-before-completion

Run verification commands and compare output against expectations before claiming completion.

Community
Advanced
vercel-labsvercel-labs
41.7k

derive-client

Generates standalone API clients from browser network traffic recorded in HAR files.

Official
Intermediate