Back to Projects
Completed Project
Browser Automation Agent
A command-line interface (CLI) tool for AI-driven browser automation.
|
RoleAI Engineer
TimelineApril 2026
TeamPersonal Project
Tech Stack4 Technologies
Mission Brief
Agent Browser is a command-line interface (CLI) tool for AI-driven browser automation. Built on top of browser-use, this tool allows you to command an AI agent to perform complex browser interactions, scraping, and testing using natural language.
Key Features
Core Capabilities
- Natural Language Control: Command the browser to navigate, click, fill forms, and extract information using simple text prompts.
- Multiple LLM Support: Easily toggle between powerful models like Google Gemini and Groq.
- Vision Mode: Support for standard screen rendering capability for complex sites.
Performance & Optimization
- Smart Rate Limiting: Built-in delay mechanism and fallback handling to gracefully circumvent free-tier API rate limits.
- CLI Workflows: Execute fast actions directly from your terminal using robust CLI commands.
- Optimized Setup: Seamless dependency management using standard virtual environment or uv.
Logging & Outputs
- Execution Traces: Application logs are automatically written to the logs/ directory for easy debugging.
- Data Exports: Any graphical outputs (like screenshots from Vision mode) or data exports are saved to the outputs/ directory.
Engineering Chronicles
API Rate Limits Exhaustion
SolutionImplemented a smart rate limiting delay mechanism and fallback handling to gracefully circumvent free-tier API restrictions (e.g., Gemini 15 RPM Free Tier).
Complex DOM Navigation
SolutionIntegrated a Vision Mode to capture screenshots and allow the LLM to process visual context for dynamically rendered sites.
Command-Line Interface Scalability
SolutionEngineered a robust CLI framework offering specific workflow executions, dynamic LLM toggling (--llm gemini), and vision toggles (--vision).
Visual Gallery
IMG_01
Installation Instructions
Clone and Setup
$git clone https://github.com/your-org/project-name.git
$cd project-name
$python -m venv venv
$venv\Scripts\activate
Install Dependencies
$pip install -e .
Environment Configuration
$cp .env.example .env
$# Fill in GOOGLE_API_KEY or GROQ_API_KEY
Run Agent
$agentbrowser run --help
$agentbrowser run "Cari jadwal libur nasional Indonesia 2026 di Google dan rangkum"
Project Access
Technologies
Python 3.11+
browser-use
Gemini
Groq
Table of Contents
- • Mission Brief
- • Key Features
- • Engineering Chronicles
- • Visual Gallery
- • Installation Instructions
Back to Projects