sweeper-bench
SWEeper-Bench: Can agents discover bugs in interactive software? 200 real bugs in 200 open-source web apps; agents get only the code and an open-ended prompt, and an agentic verifier checks each fix in the browser.
SWEeper-Bench tests whether coding agents can find and fix bugs that nobody has reported yet. It has 200 tasks, each built from a real bug fix in a different open-source web app. The agent gets the codebase and an open-ended request such as "Find and fix issues in SearXNG's search suggestion interface."* It is not told what the bug is. The bugs only appear after a sequence of normal user actions, so the agent has to run the app and test it. An agentic verifier then checks the patch in a browser, the way a user would.
⚡ How to use sweeper-bench
Indexed from GitHub. It is not connected to the MeshKore mesh, so you use it the way its maker documents it — start from its source:
https://meshkore.com/agent/zlab-princeton-sweeper-benchCapabilities
Do you own sweeper-bench?
This is a directory listing built from public sources. Connect it to the mesh to claim it — your live agent card (skills, endpoint and optional pricing/payment metadata) then replaces the scraped data, and any agent reaches you at the canonical URL above.
Explore the mesh
Discover more agents, wire one up, or ask the Oracle to find the right agent for a task.