← WingFoil Benchmark — campaign 81ebb2691f9c, execution 1

F — Knowledge & Continuity

Does the harness carry decisions and domain knowledge across fresh sessions?

S3@1.0

M-F1: earlier decisions respected or explicitly revised (higher is better)

• baseline:
  60.0% n = 1 preliminary
• baseline-docs:
  80.0% n = 1 preliminary +20.0 pp ▲ better
• openspec openspec 1.14.0 · 9cf16ae:
  100.0% n = 1 preliminary +40.0 pp ▲ better
  against openspec-docs +40.0 pp ▲ better
• openspec-docs:
  60.0% n = 1 preliminary ±0.0 pp = same
• speckit speckit v1.1.0 · f1d3a4f:
  60.0% n = 1 preliminary ±0.0 pp = same
  against speckit-docs ±0.0 pp = same
• speckit-docs:
  60.0% n = 1 preliminary ±0.0 pp = same
• wingfoil wingfoil v0.3.0 · cc31d7a:
  60.0% n = 1 preliminary ±0.0 pp = same
  against baseline-docs −20.0 pp ▼ worse

Each suite on the final snapshot

• baseline: bookings 11/11; availability 5/5; discount 4/4; hourly 10/10; cancellation 5/7
• baseline-docs: bookings 11/11; availability 5/5; discount 4/4; hourly 10/10; cancellation 7/7
• openspec openspec 1.14.0 · 9cf16ae: bookings 11/11; availability 5/5; discount 4/4; hourly 10/10; cancellation 7/7
• openspec-docs: bookings 11/11; availability 5/5; discount 4/4; hourly 10/10; cancellation 5/7
• speckit speckit v1.1.0 · f1d3a4f: bookings 11/11; availability 5/5; discount 4/4; hourly 10/10; cancellation 0/7
• speckit-docs: bookings 11/11; availability 5/5; discount 4/4; hourly 10/10; cancellation 5/7
• wingfoil wingfoil v0.3.0 · cc31d7a: bookings 11/11; availability 5/5; discount 4/4; hourly 10/10; cancellation 5/7

Hold-out

• baseline:
  hold-out 16/17 n = 1
• baseline-docs:
  hold-out 17/17 n = 1
• openspec openspec 1.14.0 · 9cf16ae:
  hold-out 17/17 n = 1
• openspec-docs:
  hold-out 16/17 n = 1
• speckit speckit v1.1.0 · f1d3a4f:
  hold-out 15/17 n = 1
• speckit-docs:
  hold-out 16/17 n = 1
• wingfoil wingfoil v0.3.0 · cc31d7a:
  hold-out 16/17 n = 1

Runs

• 81ebb2691f9c/1/runs/S3@1.0/baseline/claude-sonnet-5-5/r1 — open with bench run show 81ebb2691f9c/1/runs/S3@1.0/baseline/claude-sonnet-5-5/r1
• 81ebb2691f9c/1/runs/S3@1.0/baseline-docs/claude-sonnet-5-5/r1 — open with bench run show 81ebb2691f9c/1/runs/S3@1.0/baseline-docs/claude-sonnet-5-5/r1
• 81ebb2691f9c/1/runs/S3@1.0/openspec/claude-sonnet-5-5/r1 — open with bench run show 81ebb2691f9c/1/runs/S3@1.0/openspec/claude-sonnet-5-5/r1
• 81ebb2691f9c/1/runs/S3@1.0/openspec-docs/claude-sonnet-5-5/r1 — open with bench run show 81ebb2691f9c/1/runs/S3@1.0/openspec-docs/claude-sonnet-5-5/r1
• 81ebb2691f9c/1/runs/S3@1.0/speckit/claude-sonnet-5-5/r1 — open with bench run show 81ebb2691f9c/1/runs/S3@1.0/speckit/claude-sonnet-5-5/r1
• 81ebb2691f9c/1/runs/S3@1.0/speckit-docs/claude-sonnet-5-5/r1 — open with bench run show 81ebb2691f9c/1/runs/S3@1.0/speckit-docs/claude-sonnet-5-5/r1
• 81ebb2691f9c/1/runs/S3@1.0/wingfoil/claude-sonnet-5-5/r1 — open with bench run show 81ebb2691f9c/1/runs/S3@1.0/wingfoil/claude-sonnet-5-5/r1

Scenarios that also touch this category

• S2@1.0: primary category D

Other models (slices)

Reported apart from the campaign’s model, and never in its comparisons.

• baseline, claude-haiku-5-5: M-Q1 35/37 (n = 1)
• wingfoil wingfoil v0.3.0 · cc31d7a, claude-haiku-5-5: M-Q1 30/37 (n = 1)