skills.rest
Wizarck
WizarckCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

agent-browser

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

2|Updated Apr 19, 2026
Search Tags:#screenshots#cli#browser automation#web scraping#form filling#e2e testing#chrome cdp
One-click install
npx skills add https://github.com/Wizarck/nexandro --skill agent-browser-wizarck
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/Wizarck/nexandro/tree/main/skills/agent-browser
Command: npx skills add https://github.com/Wizarck/nexandro --skill agent-browser-wizarck

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires agent-browser, and includes references (resource) components.

What problem does it solve? AI agents and automation scripts need a reliable way to interact with websites—navigating pages, filling forms, clicking buttons, and extracting data—without parsing raw HTML or managing complex browser drivers. This Skill provides a CLI that controls Chrome/Chromium via CDP with compact element refs that minimize token usage. ## Core Features & Use Cases - Snapshot-Based Interaction: Get interactive element refs (@e1, @e2) from accessibility-tree snapshots, then click, fill, select, and type using those refs instead of brittle CSS selectors. - Authentication & Sessions: Reuse login state via auth vault, persistent profiles, session names, or state files, with support for OAuth, 2FA, and encrypted credential storage. - Capture & Verification: Take screenshots (including annotated and full-page), save PDFs, record video, diff page states, and inspect network requests for testing and debugging. - Use Case: Automate an end-to-end web app test—open the login page, authenticate with a saved profile, fill a multi-field form, submit it, and capture an annotated screenshot verifying the result. ## Quick Start Ask the agent to open a website, snapshot the page, fill in the login form, and take a screenshot of the result.

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 agent-browser?▼

Install agent-browser via npm, brew, or cargo, then run `agent-browser open <url>` followed by `agent-browser snapshot -i` to get element refs. Use those refs with commands like `click`, `fill`, and `screenshot` to interact with the page.

How do I fill out and submit a web form programmatically?▼

Navigate to the form page, run `snapshot -i` to discover input refs, then use `fill @e1 "value"` for text fields, `select` for dropdowns, and `check` for checkboxes. Finish by clicking the submit button ref and waiting for the result page.

How do I handle login and authentication in browser automation?▼

Use the auth vault (`agent-browser auth save` / `auth login`) to store encrypted credentials, or reuse an existing Chrome profile with `--profile`. You can also save and load session state files, or import cookies from a running Chrome via `--auto-connect`.

Why do I get 'ref not found' errors after clicking an element?▼

Element refs are invalidated whenever the page changes, such as after navigation, form submission, or dynamic content updates. Re-run `agent-browser snapshot -i` after any page change to obtain fresh refs before interacting again.

Can agent-browser run parallel or isolated browser sessions?▼

Yes, use the `--session <name>` flag to create isolated browser contexts with separate cookies, storage, and tabs. This supports concurrent scraping, A/B testing, and running multiple agents without conflicts.

Does agent-browser support mobile or iOS testing?▼

Yes, it supports device emulation via `set device "iPhone 14"` for viewport and user agent spoofing. It can also drive real iOS Simulator Safari sessions using `-p ios` on macOS with Xcode and Appium installed.

Related Skills

Explore More Software Engineering Skills→
MantisWareMantisWare

import

Import workspace templates from local files, URLs, or GitHub repositories with collision handling and secret prompts.

Community
Advanced
rafaelcostaf4-afkrafaelcostaf4-afk

playwright-cli

Automate browser interactions for testing, form filling, and data capture.

Community
Intermediate
abretonc7sabretonc7s

visual-testing

Capture screenshots before and after UI interactions to detect visual regressions.

Community
Intermediate
magnus919magnus919
40

flaresolverr-cli

Execute browser-backed HTTP requests to bypass Cloudflare and DDoS-GUARD anti-bot protections.

Community
Intermediate