skills.rest
HyuseCS
HyuseCSCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

vc-agent-browser

Automate browser navigation, interaction, and testing via CLI with context-efficient snapshots.

Updated Apr 9, 2026
Search Tags:#browser automation#web testing#playwright#puppeteer#screenshot#headless browser#browserbase
One-click install
npx skills add https://github.com/HyuseCS/project-c --skill vc-agent-browser-hyusecs
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: vc-agent-browser
Source: https://github.com/HyuseCS/project-c/tree/main/.claude/skills/vc-agent-browser
Command: npx skills add https://github.com/HyuseCS/project-c --skill vc-agent-browser-hyusecs

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Browser automation for AI agents typically consumes excessive context tokens and requires verbose tooling. This Skill provides a CLI-driven browser automation workflow using compact accessibility snapshots with element refs, reducing context usage by roughly 93% compared to Playwright MCP while supporting long autonomous sessions, video recording, and cloud browsers. ## Core Features & Use Cases - Snapshot + Refs Workflow: Navigate pages, capture interactive element snapshots with stable @refs, then click, fill, and interact using those refs. - Full Browser Control: Screenshots, PDF export, video recording, network interception, cookie/storage management, multi-tab handling, and device emulation. - Cloud Browser Support: Native Browserbase integration for CI/CD pipelines and environments without a local browser. - Puppeteer Script Utilities: Companion scripts for auth injection, cookie import, console/network monitoring, WebSocket debugging, and ARIA snapshots. - Use Case: Run a self-verifying test loop where an agent opens a login page, snapshots the form, fills credentials via refs, submits, waits for the dashboard URL, and records a video of the session for debugging. ## Quick Start Ask the agent to open a URL with agent-browser, take an interactive snapshot, and click or fill elements using the returned refs.

Frequently Asked Questions about vc-agent-browser

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

FAQPage Schema
How do I automate browser interactions with agent-browser?▼

Follow the four-step pattern: open a URL, run snapshot -i to get interactive elements with @refs, interact using commands like click @e1 or fill @e2, then re-snapshot after page changes. Install globally with npm install -g agent-browser first.

What is the difference between agent-browser and chrome-devtools scripts?▼

agent-browser is a Rust-based CLI suited for long autonomous sessions, video recording, and Browserbase cloud browsers. The chrome-devtools Puppeteer scripts offer custom utilities like WebSocket full-frame debugging, automatic screenshot compression, and fine-grained console filtering.

Can I use agent-browser with cloud browsers in CI/CD?▼

Yes, agent-browser supports Browserbase natively. Set BROWSERBASE_API_KEY and BROWSERBASE_PROJECT_ID environment variables, then run commands with the -p browserbase flag to execute against remote cloud browsers in pipelines like GitHub Actions.

How do I test authenticated pages with browser automation scripts?▼

Use inject-auth.js to inject cookies, Bearer tokens, or localStorage/sessionStorage items into the session, or import-cookies.js to load cookies exported from browser extensions. Auth persists in .auth-session.json for 24 hours across script runs.

Why is my element not found after clicking in agent-browser?▼

Element refs become stale after page changes or navigation. Re-run snapshot -i after any interaction that modifies the DOM to get fresh refs before continuing interactions.

Does agent-browser support video recording of sessions?▼

Yes, use record start before performing actions and record stop to save the session as a .webm video file. This is useful for debugging autonomous workflows and verifying test behavior.

Related Skills

Explore More Software Engineering Skills→
robertzengcnrobertzengcn

setup-browser-cookies

Import cookies from local Chromium browsers into headless sessions for authenticated testing.

Community
Advanced
gonzoblascogonzoblasco
1

e2e-testing-patterns

Implements end-to-end test suites with stable selectors, retries, and CI parallelization.

Community
Intermediate
nathanvalenathanvale
6

observability

Implements structured JSONL logging, correlation IDs, and performance metrics for Bun applications.

Community
Advanced
HouseGarofaloHouseGarofalo
3

browserless

Automate headless Chrome browser tasks via REST API and WebSocket.

Official
Intermediate