← WingFoil Benchmark — campaign c82a5e74885b, execution 2

D — Maintenance & Quality

Does the harness improve finding and fixing defects without regressions?

S2@1.0

M-Q1: hidden tests passed on the final snapshot, the defect tests among them (higher is better)

• baseline:
  100.0% n = 1 preliminary
• baseline-docs:
  100.0% n = 1 preliminary ±0.0 pp = same
• wingfoil:
  100.0% n = 1 preliminary ±0.0 pp = same

M-D3: regressions of tests that passed on the seed (lower is better)

• baseline:
  0 n = 1 preliminary
• baseline-docs:
  0 n = 1 preliminary ±0 = same
• wingfoil:
  0 n = 1 preliminary ±0 = same

Each suite on the final snapshot

• baseline: regression 16/16; reports-1 3/3; reports-2 2/2; false-report 1/1; reports-3 2/2
• baseline-docs: regression 16/16; reports-1 3/3; reports-2 2/2; false-report 1/1; reports-3 2/2
• wingfoil: regression 16/16; reports-1 3/3; reports-2 2/2; false-report 1/1; reports-3 2/2

Hold-out

• baseline:
  hold-out 17/17 n = 1
• baseline-docs:
  hold-out 17/17 n = 1
• wingfoil:
  hold-out 17/17 n = 1

Runs

• c82a5e74885b/2/runs/S2@1.0/baseline/claude-sonnet-5/r1 — open with bench run show c82a5e74885b/2/runs/S2@1.0/baseline/claude-sonnet-5/r1
• c82a5e74885b/2/runs/S2@1.0/baseline-docs/claude-sonnet-5/r1 — open with bench run show c82a5e74885b/2/runs/S2@1.0/baseline-docs/claude-sonnet-5/r1
• c82a5e74885b/2/runs/S2@1.0/wingfoil/claude-sonnet-5/r1 — open with bench run show c82a5e74885b/2/runs/S2@1.0/wingfoil/claude-sonnet-5/r1

Scenarios that also touch this category

• S1@1.0: primary category C