BigHugger
sk Skill · majiayu000

eval-harness

Assessment-driven development — Quantify code generation quality with pass@k / pass^k metrics, automatically scored by Grader.

installs 8w
0
30-day movement
starts with the next reading
Related entries
12
Connections
0
Shell
Host repository
majiayu000/vibeguard
Host stars
41
Host language
Shell