PulsarRPA (Browser4)

Category: OSS Agent / Browser Automation Engine

License: Apache 2.0 (free, open source)

Official repository: https://github.com/platonai/PulsarRPA

Companion project (Auto Extraction): https://github.com/platonai/PulsarRPAPro

---

What It Is

PulsarRPA (branded as Browser4) is a lightning-fast, coroutine-safe browser engine for AI agents. It is written in Kotlin and runs on the JVM. It wraps Chrome via CDP and exposes a high-level session API for workflow automation, parallel page crawling, and hybrid data extraction combining LLM calls, ML models, and CSS selectors.

Key capabilities:

• Browser Automation -- high-performance automation for navigation, interaction, and data extraction with full CDP control.
• Extreme Throughput -- coroutine-safe architecture supports 100k to 200k complex page visits per machine per day.
• Hybrid Data Extraction -- mix LLM extraction, ML-based auto-extraction, and CSS/XPath selectors in a single pipeline.
• X-SQL -- SQL-like query language that combines DOM selectors with LLM functions for structured extraction across dozens of entities and hundreds of fields.

Free vs Paid

PulsarRPA (Browser4) is free and open source under the Apache 2.0 license. The Auto Extraction ML feature is available via the companion jar PulsarRPAPro (also free). No paid tier or SaaS subscription exists as of June 2026.

Safety and ToS Notes

PulsarRPA includes built-in proxy rotation and block-resistant retry logic designed for unblocking website access at scale. Users are responsible for complying with the terms of service of any target website. The library itself carries no usage restrictions beyond Apache 2.0, but scraping a site in violation of its ToS is the operator's liability.