LaunchKit · 2026
Back to Skills

agentic-browser-0-1-2

Browser automation for AI agents via inference.sh.

0
906 downloads
by @xyny89

Setup & Installation

openclaw skills install @xyny89/agentic-browser-0-1-2

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install agentic-browser-0-1-2

What This Skill Does

Browser automation for AI agents via inference.sh. Uses Playwright under the hood with an @e ref system that assigns short identifiers to interactive page elements. Supports clicking, form filling, drag-and-drop, file upload, JavaScript execution, screenshots, and video recording.

The @e ref system abstracts away CSS selectors and XPaths, making element targeting simpler to use in agent workflows without managing a local browser environment.

When to use it

  • Auto-filling and submitting web forms on behalf of a user
  • Scraping product prices or listings from e-commerce pages
  • Recording browser sessions to debug UI regressions
  • Logging into sites and extracting data from authenticated pages
  • Running scripted browser workflows from the command line

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: search Google for 'weather today' and return the page snapshot

AGENT
  1. 1Open a new browser session at https://google.com
  2. 2Receive @e refs for interactive elements including the search input
  3. 3Fill the search input with 'weather today' and press Enter
  4. 4Wait for results to load, then call snapshot to get fresh @e refs
  5. 5Close the session and return the snapshot text
OUTPUT

Page snapshot with element refs showing weather search results

Requirements

Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.

inference.sh accountinference.sh CLI (infsh) installed