browser-oxide 0.1.0

BrowserOxide — stealth headless browser engine in Rust (Python bindings) — real HTML/CSS/DOM/JS, own BoringSSL TLS, native fingerprint, no Chromium

BrowserOxide (Python) — browser-oxide on PyPI

A stealth headless browser engine, in Rust, with Python bindings. A real HTML/CSS/DOM/JavaScript browser built from scratch — own BoringSSL TLS stack, native fingerprint, V8 JavaScript — with no Chromium, no CDP driver, no Selenium underneath. Built for scraping, archival, and AI-agent web access against modern anti-bot stacks (Cloudflare, Akamai, AWS WAF, DataDome).

pip install browser-oxide

from browser_oxide import Browser, Profile, Verdict

with Browser(profile=Profile.chrome()) as b:
    page = b.navigate("https://example.com")
    print(page.title, len(page.html), page.verdict)
    if page.verdict == Verdict.PASS:           # challenge pages return HTTP 200 —
        print(page.evaluate("navigator.userAgent"))   # trust the verdict, not status
        print(page.query_text("h1"))

API

Profile.chrome() / .firefox() / .iphone() / .pixel()	built-in identities (TLS + headers + navigator + GPU, coherent)
Profile.from_file(path)	custom profile (YAML/JSON)
Browser(profile=None)	spawns the engine; context-manager friendly
b.navigate(url, max_iterations=5) -> Page	cold navigation (GIL released)
page.url / .title / .html / .text / .verdict / .is_challenge	properties
page.evaluate(js) -> str · page.query_text(sel) -> str | None	scripting
Verdict.PASS / THIN_SHELL / EDGE_BLOCK / SENSOR_FAIL / …	honest render outcome

What's real vs. not

Real: from-scratch TLS (JA3/JA4-accurate), full V8, real DOM/CSS, Canvas 2D raster, 100+ coherent fingerprint properties. Not implemented: real WebGL raster, HTTP/3 by default, built-in per-vendor bypass (Kasada is the open gap). Full detail + measured pass rates: see the repo docs (getting-started-python.md, guides/STEALTH_FAQ.md, BENCHMARK.md).

MIT OR Apache-2.0.