skills.rest
tmtgroupbot
tmtgroupbotCOMMUNITY
Explore all skills by this author →
GitHub

Agent Browser

Automates headless browser navigation, interaction, and page snapshots via CLI commands.

Updated Mar 30, 2026
Search Tags:#cli#browser automation#web scraping#form filling#playwright#ui testing#headless browser
One-click install
npx skills add https://github.com/tmtgroupbot/lptracker-deploy --skill agent-browser-tmtgroupbot
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Agent Browser
Source: https://github.com/tmtgroupbot/lptracker-deploy/tree/main/skills/agent-browser
Command: npx skills add https://github.com/tmtgroupbot/lptracker-deploy --skill agent-browser-tmtgroupbot

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires agent-browser.

What problem does it solve? AI agents and developers need a programmatic way to browse the web—navigating pages, clicking elements, filling forms, and extracting data—without writing custom Playwright scripts for every task. ## Core Features & Use Cases - Structured Page Interaction: Snapshot pages to get interactive elements with stable refs, then click, fill, type, and drag using those refs. - Data Extraction & Capture: Extract text, attributes, screenshots, PDFs, and video recordings from any page. - Session & State Management: Save and load authentication state, manage cookies/storage, and run parallel isolated browser sessions. - Use Case: Log into a web app once, save the session state to a file, then reuse it in later runs to scrape dashboard data without re-authenticating. ## Quick Start Use the agent-browser skill to open https://example.com, take an interactive snapshot, and fill in the login form.

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?▼

Install agent-browser via npm, then use commands like open, snapshot -i, click, and fill to navigate pages and interact with elements. Snapshots return refs like @e1 that you pass to interaction commands.

How to fill and submit a web form programmatically?▼

Open the page, run snapshot -i to get element refs, then use fill @ref "value" for each input and click on the submit button ref. Add a wait command after submission to handle page loads.

Does agent-browser support saving login sessions?▼

Yes, agent-browser supports session persistence through state save and state load commands. Save cookies and storage to a JSON file after logging in, then load it in later sessions to skip re-authentication.

Why are my element refs not working after navigation?▼

Refs are stable only per page load and become invalid after navigation or major DOM changes. Always run snapshot again after navigating to get fresh refs before interacting with elements.

Can agent-browser run multiple browsers in parallel?▼

Yes, use the --session flag with a unique name to run isolated browser sessions in parallel. Each session maintains its own cookies, storage, and tabs independently.

Related Skills

Explore More Software Engineering Skills→
graysurfgraysurf

browser-session

Record browser-session goals, steps, artifacts, and verification status via nils-cli.

Community
Intermediate
abhishekmmgnabhishekmmgn

scrape

Extract text, tables, and metadata from web pages via Bash.

Community
Advanced
GabeujinGabeujin

chrome-devtools

Automate browser interactions and analyze network traffic with Chrome DevTools.

Community
Advanced
HouseGarofaloHouseGarofalo
3

browserless

Automate headless Chrome browser tasks via REST API and WebSocket.

Official
Intermediate