HN Showcase
Back to browse
GitHub Repository
2 starsRust

Webact – token-efficient browser control for AI agents (GitHub)

by kxbnb·Mar 3, 2026·1 point·1 comment
Visit ProjectView on HN

AI Analysis

●●●BangerSolve My ProblemBig Brain

Slashes tokens vs Playwright by returning a DOM brief instead of full tree.

Strengths
  • •DevTools Protocol instead of Playwright eliminates bloat and dependencies.
  • •Single JS file, zero deps, integrates with 40+ agent tools via skills spec.
  • •Session isolation lets multiple agents run in same Chrome without stepping on each other.
Weaknesses
  • •Visual pages still need screenshots, doesn't fully eliminate fallback costs.
  • •Narrow niche—only useful if you're already building LLM browser agents.
Target Audience

AI agent builders, LLM application developers

Similar To

Playwright · Pupeteer · Selenium

Post Description

I built webact while experimenting with browser control for LLM agents and kept running into the same problem: token usage.

Most tools wrap Playwright and return huge accessibility trees or DOM dumps, which gets expensive quickly. webact takes a simpler approach — it talks directly to Chrome via the DevTools Protocol and returns a small "page brief" instead.

It’s just a single JS file (~196KB), zero dependencies, and it uses your existing Chrome session.

Happy to hear thoughts.

Similar Projects