ClawBench

Open-source benchmark that tests whether AI browser agents can complete everyday online tasks.

PlatformWebfreeglobal

ClawBench is an open-source benchmark that tests how well AI browser agents complete everyday online tasks, like booking travel or applying for a job, across live websites, and publishes a public leaderboard of results. It ships as an installable command-line tool that runs an agent in an isolated container and scores it against a human reference, aimed at teams and researchers who need an objective way to evaluate a browser agent's real-world reliability.

Website

Full match profile

Behind the summary, Matchbox keeps a richer profile of ClawBench - the signals our matcher actually reads to decide when to surface it. It stays private; claim the listing to see and control it.

Problem & pain-point mapping
Who we surface it to (audience fit)
What it's a strong alternative to
Trust & credibility signals

Claim ClawBench to unlock →

Something wrong with this listing — dead link, not a real product, wrong info?

Try Matchbox with your own problem

Describe what is not working - we’ll show you whether ClawBench (or something else) actually fits.

Try a problem