skills.rest
aimentor606
aimentor606COMMUNITY
Explore all skills by this author →
GitHub

agent-browser

Automate browser navigation, form filling, and data extraction via CLI.

Updated Apr 8, 2026
Search Tags:#automation#cli#ai-agent#browser#data-extraction#web-automation
One-click install
npx skills add https://github.com/aimentor606/aether --skill agent-browser-aimentor606
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/aimentor606/aether/tree/main/core/kortix-master/opencode/skills/KORTIX-system/agent-browser
Command: npx skills add https://github.com/aimentor606/aether --skill agent-browser-aimentor606

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

AI agents often need to interact with web pages to perform tasks such as logging in, filling forms, or extracting data. Traditional approaches require manual scripting or human-in-the-loop steps; this Skill provides a deterministic, scriptable browser automation layer that preserves context for decision-making.

Core Features & Use Cases

  • Deterministic browser control via CLI: open pages, snapshot elements, and interact using refs.
  • Snapshot-based interactions: use @eN refs to drive clicks, fills, and navigation without brittle selectors.
  • Session isolation and multi-page workflows: run concurrent automations with separate cookies, storage, and state.
  • Data capture and validation: extract text, URLs, and forms data, then export or verify results.

Quick Start

Open a target website, capture interactive element references, and run scripted actions to automate a browser task.

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 tasks with AI agents without using brittle CSS selectors?▼

AI agent browser automation uses snapshot-based element references like @eN to drive clicks and form filling. This approach replaces brittle CSS selectors with interactive element refs captured from page snapshots, ensuring reliable navigation and data extraction workflows.

Can I run concurrent web automation workflows with isolated sessions and cookies?▼

Yes, web automation supports session isolation for concurrent workflows across multiple pages. You can run parallel browser automations with separate cookies, local storage, and state, enabling multi-page data extraction without cross-session interference.

What is the best way to extract web data and form inputs using CLI browser automation?▼

CLI browser automation captures text, URLs, and form data from web pages using deterministic command patterns. You open target pages, snapshot interactive elements, and script actions to extract and validate website data for export.

Does deterministic browser control require a specific runtime environment to function?▼

Deterministic browser control requires Chromium-based runtimes via the agent-browser interface. This environment supports scriptable website navigation, form filling, and data extraction while preserving context for AI agent decision-making.

How do I fill out web forms automatically using a scriptable browser layer?▼

You fill out web forms by opening the target page, capturing a snapshot of interactive elements, and using element refs to script deterministic inputs. This scriptable browser layer automates form filling without manual human-in-the-loop steps.

What are the limitations of using snapshot-based element references for web automation?▼

Snapshot-based element references require capturing page state before each interaction, meaning highly dynamic pages may need frequent snapshot refreshes. This approach ensures reliable automation but depends on Chromium-based runtimes for consistent execution.

Related Skills

Explore More Software Engineering Skills→
davekilleendavekilleen
461

agent-browser

Automate browser tasks via CLI using accessibility snapshots and ref-based element selection.

Community
Intermediate
mupengi-botmupengi-bot
11

data-scraper

Fetch HTML from URLs and extract structured data into JSON or Markdown.

Community
Intermediate
traygerbigtraygerbig
1

Brain Trust

Orchestrate multi-agent teams with role delegation and consensus protocols.

Community
Advanced
vm0-aivm0-ai
1.1k

agent-browser

Automate browser navigation, form filling, screenshots, and data extraction.

Official
Intermediate