Mochi.js

What it does

Mochi.js is a browser automation library built for Bun that uses the Chrome DevTools Protocol to control Chromium. It handles programmatic browsing tasks while attempting to avoid detection by anti-bot systems. The library generates fingerprints—browser characteristics like user agent, canvas rendering, WebGL output, and audio context—from a consistent seed so they remain coherent across probes. It simulates human behavior through click, typing, and scroll patterns based on biomechanical models. It can auto-solve Turnstile challenges and inject custom code into pages before scripts run.

Who it is for

Developers building automation workflows who need to avoid fingerprint detection. The library targets users currently relying on Playwright or Puppeteer but facing blocking from detection systems. It suits those performing bot-protected tasks at scale: scraping, testing, account access, or checkout automation on sites with strict anti-bot measures.

Pricing

The site does not show prices.

How it stands out

Mochi.js claims to differ from mainstream tools in three ways. First, it uses raw CDP instead of higher-level abstractions, giving lower-level control. Second, all fingerprint surfaces—canvas, fonts, WebGL, audio, device media lists—derive from a single seed through a 48-rule system, preventing mismatched attributes. Third, it routes HTTP requests through Chromium itself rather than a parallel HTTP layer, ensuring TLS signatures and HTTP/2 behavior match real Chrome. The library is Bun-native with no Node or Python dependencies. It includes a test harness comparing live sessions against captured baselines from real devices, blocking commits that deviate without explanation.

What a founder should check

Three areas warrant verification before building a competitor. First, how quickly do detection systems adapt to this approach? If fingerprint-based detection is already bypassed by many tools, the moat may erode faster than assumed. Second, what switching costs exist for users of Playwright or Puppeteer? Rewriting automation code for a new library has friction; understand whether that barrier protects adoption. Third, examine whether Chromium-native fetch truly prevents detection or simply delays it. If sites move beyond fingerprinting to behavioral and network-level anomaly detection, the claimed advantage shrinks.