skills.rest
iterate
iterateOFFICIAL
Explore all skills by this author →
Website𝕏@iterateGitHub

agent-browser

Automate Chrome browser interactions via CDP with snapshot-based element refs.

198|19|Updated Sep 15, 2025
Search Tags:#authentication#screenshots#browser automation#web scraping#ui testing#CDP#element references
One-click install
npx skills add https://github.com/iterate/iterate --skill agent-browser-iterate
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/iterate/iterate/tree/main/.agents/skills/agent-browser
Command: npx skills add https://github.com/iterate/iterate --skill agent-browser-iterate

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It solves the problem of reliably interacting with complex websites—navigation, login, form filling, clicks, scraping, and verification—by converting the page into compact, AI-actionable element references.

Core Features & Use Cases

  • Deterministic browser automation: Navigate pages and perform actions using Chrome/Chromium driven by CDP.
  • Snapshot-to-ref workflow: Take snapshots to obtain stable element refs, then interact, re-snapshot, and diff/verify outcomes.
  • Auth and evidence capture: Import or persist sessions, and capture screenshots/PDFs and optional video for debugging and auditing.
  • Use Case: Automate a protected web workflow by restoring saved auth state, submitting a form, waiting for results, extracting page text, and saving a screenshot for evidence.

Quick Start

Use the browser automation workflow to open a page, take an interactive snapshot, then click or fill the referenced elements.

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How do I automate web scraping and form filling with AI agents?▼

AI agents automate browser interactions by converting complex website DOMs into compact, AI-actionable element references. This enables reliable navigation, form filling, and data extraction using a snapshot-and-ref lifecycle driven by Chrome/Chromium CDP.

How does browser automation handle dynamic DOM changes and iframes?▼

Browser automation handles dynamic DOM changes and iframes using a snapshot-to-ref workflow. It takes interactive snapshots to obtain stable element references, performs actions, then re-snapshots and diffs the page to verify outcomes and maintain reliability despite dynamic updates.

Can I use saved authentication sessions for UI testing and web scraping?▼

Yes, you can import or persist saved authentication sessions for UI testing and web scraping. The automation restores auth state to access protected web workflows, submit forms, wait for results, and capture screenshots or PDFs for evidence.

What's the best way to capture screenshots and PDFs during E2E testing?▼

The best way to capture screenshots and PDFs during E2E testing is using Chrome/Chromium CDP automation. It supports capturing screenshots, PDFs, and optional video for debugging and auditing, while extracting page text and verifying UI outcomes through snapshot diffs.

Does browser automation require Chrome or Chromium CDP control to work?▼

Yes, deterministic browser automation requires Chrome or Chromium CDP control to function. This protocol control enables the snapshot-and-ref lifecycle, session persistence, domain safety controls, and extraction of JSON outputs for AI-driven web interactions.

What are the limitations of using snapshot-and-ref workflows for web automation?▼

The snapshot-and-ref workflow requires re-snapshotting after actions to verify outcomes, which adds overhead to web automation. If a page's dynamic DOM changes too rapidly between snapshots, element references may become stale and require additional diff cycles to maintain reliability.

Related Skills

Explore More Software Engineering Skills→
npc-livenpc-live
168

agent-browser

Automate browser interactions via Chrome/CDP using the agent-browser CLI.

Official
Intermediate
alibabaalibaba
10.5k

pdf-extractor

Extract text, tables, and metadata from PDF documents into JSON.

Official
Intermediate
skaile-aiskaile-ai

agent-browser

Automate browser navigation, form interaction, and data extraction for AI agents.

Official
Advanced
martinjlowmmartinjlowm

agent-browser

Automates browser interactions including navigation, clicking, filling, waiting, and screenshots for web UI testing and data extraction.

Community
Intermediate