browser-agent-server

Deploy a FastAPI browser automation server with VNC streaming and dashboards.

19|6|Updated Feb 4, 2026
One-click install
npx skills add https://github.com/niveshdandyan/happycapy-browser-agent --skill browser-agent-server
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browser-agent-server
Source: https://github.com/niveshdandyan/happycapy-browser-agent/tree/main
Command: npx skills add https://github.com/niveshdandyan/happycapy-browser-agent --skill browser-agent-server

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires fastapi, uvicorn, websockets, websockify, pydantic, playwright, browser-use, and includes scripts (resource) components.

What problem does it solve?

Deploy a full-stack Browser Agent server that federates a FastAPI backend, real-time dashboard, Xvfb/VNC rendering, and multi-model LLM strategies to automate browser tasks with governance and live monitoring.

Core Features & Use Cases

  • FastAPI backend provides a robust API surface to start tasks, monitor status, and stream live updates.
  • Live VNC streaming allows real-time observation of a browser session as tasks execute.
  • Council Planning & multi-model strategies supports planner-executor, single, fallback_chain, consensus, and council modes for reliability, quality, and anti-stall capabilities.
  • Single-file dashboard is a self-contained UI with no build step, accessible from a browser.
  • Flexible deployment suitable for on-premises or cloud environments to automate complex browser workflows.

Quick Start

Install dependencies, export your AI gateway API key, start the server, and then open the dashboard to run a task.

Frequently Asked Questions about browser-agent-server

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

FAQPage Schema
How do I deploy a multi-model browser automation server with FastAPI?

Deploy a multi-model browser automation server by running a FastAPI backend with Uvicorn, configuring your LLM API keys, and starting the server to orchestrate browser tasks with live WebSocket updates.

What is council planning in multi-model browser automation?

Council planning is a multi-model governance strategy for browser automation that uses consensus, fallback chains, or planner-executor modes to improve task reliability and prevent stalling.

Can I stream live browser execution sessions using VNC and WebSocket?

Yes, you can stream live browser execution sessions using Xvfb and VNC rendering alongside WebSocket events to observe and monitor automated browser tasks in real-time from a dashboard.

Does this browser automation server require a frontend build step for the dashboard?

No, the browser automation server provides a single-file, self-contained live dashboard with no build step required, making it accessible directly from a browser after starting the server.

What's the best way to govern AI web automation workflows to prevent stalls?

The best way to govern AI web automation workflows is using multi-model strategies like fallback chains and consensus modes, which provide anti-stall capabilities and ensure reliable task execution.