Websites · PROJECT NOTES
WebBench
A public collection of live-website task instructions spanning reading, account changes and file operations, accompanied by published evaluation results.
What can you do with it?
Select concrete browser workflows by action category and starting URL, then use a separately supplied authorized agent and outcome-review procedure.
What should you know before using it?
The repository publishes CSV tasks and results, not a complete resettable environment or evaluator package. WRITE tasks change real account state and need appropriate test-account authorization. The current main CSV has 2,647 records while the README reports 2,454; pin the file. All 315 HITL rows also occur in the main collection.
What is verified here?
Runtime: Bring your own browser agent, accounts where needed and task-outcome assessment. Public fields are ID, Starting URL, Category and Task; no executable outcome tests are embedded. Sites, authentication and downloads can change. No live task was attempted.
License note: Repository LICENSE and official Hub card declare MIT. This does not cover third-party website content or authorize account mutations. Upstream license ↗
This entry was reviewed against original project material. We have not independently run the software or reproduced its results. AI-development claims are author-reported; check upstream for current versions, setup requirements, and licenses.
- Resource type
- Dataset + benchmark
- Provenance
- Development provenance not established
- Reviewed
- Platforms
- Web