Real-world browser-agent benchmark: 210 tasks across 107 websites, multi-agent/multi-browser evaluation, reproducible leaderboard and result submissions.
Latest release v1.0.0 — browseruse-bench v1.0.0 · 11 May 2026
3 files for GitHub Copilot, Codex, OpenCode and Claude Code: browseruse-agent-bench copilot-instructions.md, browseruse-agent-bench AGENTS.md, browseruse-agent-bench CLAUDE.md — 4,862 tokens loaded in every session.
.github/copilot-instructions.md A 3,057 tok AGENTS.md A 386 tok CLAUDE.md A 1,419 tok These files are lexmount/browseruse-agent-bench's own configuration — they tell GitHub Copilot, Codex, OpenCode and Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.