Mahal Systems

Overview › Benchmark catalog

FEA-Bench

Fits incremental-development agents in Python; the task format (issue plus new component signatures plus tests) is a usable template for feature cases.

What it asks the agent to do

  • add a feature to a repository given a description and the new function signatures
  • make new unit tests from the feature pull request pass

What to borrow from it

case template

Caveats

Instance count and data license were not visible on the paper page; the code link is not on the arXiv page, so check availability before planning on it.

Details

Workloadscoding
Interactionautonomous long horizon
Environmentcontainers
End state checkedYes
Toolsshell, file editor
Gradingunit tests
SizeNot published
Datareal public
Personal datapossible
Licenseunknown
Harnessown repo
Statusactive

Links