Microsoft's WebHarbor docks 153 real websites into dockerized environments for GUI agent RL
microsoft · hf · 2026-11-04
Microsoft released WebHarbor, providing executable environments for GUI agent evaluation and RL training.
- Dockerizes 153 real websites into 3,518 tasks, ready for reinforcement learning
- Supports multi-step and post-authentication tasks with stable, reproducible environments
- Built by 57 contributors from 26 institutions (Stanford, Meta, NUS, etc.)
- Training in WebHarbor yields strong generalization on multi-step cross-site tasks
Paper: arXiv 2611.03634; code: github.com/aiming-lab/WebHarbor
More from coding & agent
- EvoTeam evolves multi-agent team organization and skills from execution experience — JLU-Deep · 2026-11-04
- Switch-SWE benchmark shows coding agents lose focus when switching unfinished tasks — SJTU · 2026-11-04
- Fan builds draggable "Tibo calibrator" meme page with Nano Banana and RIFE for $0.24 — GlobalProcedure1731 · 2026-11-04
- Valid tokens, allowed actions, zero intent checks: the agent security gap nobody solves — Cytersmyz · 2026-11-04
- Stanford CS329X HW3: Coco agent traces plus AutoMetrics evaluators, open-sourced — EchoShao8899 · 2026-11-04
- Netflix shares agentic automation for auto-diagnosing failed data and ML workloads — AxSaucedo · 2026-11-04