All workflows

Workflow runs

• Add program scope enforcement as a preflight gate

  failed:
  ci#29:Commit5a5effbpushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 29

• Negation detection: deterministic structure closes the case again

  failed:
  ci#28:Commitfcc797bpushedbyZaidux
  main

  Oct 9, 20264s
  More actions for workflow run 28

• Measure the reasoning gap: refutes the simple case for a model

  failed:
  ci#27:Commit36d9ee9pushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 27

• Add spec-derived state inference; declared auth transport overrides p…

  failed:
  ci#26:Commit7306755pushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 26

• Widen coverage with recon from published API specifications

  failed:
  ci#25:Commitbb495d8pushedbyZaidux
  main

  Oct 9, 202640s
  More actions for workflow run 25

• Fix all five work-queue items; over-flag 0.037 to 0.000, recall 0.60 …

  failed:
  ci#24:Commit851a2b3pushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 24

• First independent applicability measurement: over-flag rate 0.037, re…

  failed:
  ci#23:Commitfe1bf12pushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 23

• Add isolated judge material and the scorer for independent judgements

  failed:
  ci#22:Commit1231406pushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 22

• Add annotation form; fix six fact-inference bugs it exposed

  failed:
  ci#21:Commit658d3f3pushedbyZaidux
  main

  Oct 9, 20265s
  More actions for workflow run 21

• Measure over-flagging against BenchmarkJava: cannot, and here is why

  failed:
  ci#20:Commit37f66f2pushedbyZaidux
  main

  Oct 9, 20264s
  More actions for workflow run 20

• Add real target sources: BenchmarkJava negatives and Juice Shop

  failed:
  ci#19:Commite0d4a38pushedbyZaidux
  main

  Oct 9, 20264s
  More actions for workflow run 19

• Add real-target acquisition machinery

  failed:
  ci#18:Commit5430852pushedbyZaidux
  main

  Oct 9, 20264s
  More actions for workflow run 18

• Add labelled-set format, scaffolding, and annotation guide

  failed:
  ci#17:Commitec5ec2apushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 17

• Wire four-dimension feedback into the paired A/B harness

  failed:
  ci#16:Commit1eb78c1pushedbyZaidux
  main

  Oct 9, 202640s
  More actions for workflow run 16

• Implement four-dimension coverage scoring

  failed:
  ci#15:Commitbd058dbpushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 15

• Assess external reviews; fix the tool-name inference defect

  failed:
  ci#14:Commit9e21b94pushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 14

• Widen gate with NVD; reject CyberGym on evidence

  failed:
  ci#13:Commit9a4102epushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 13

• Rewrite docs around the scorer architecture

  failed:
  ci#12:Commit04e303fpushedbyZaidux
  main

  Oct 9, 20264s
  More actions for workflow run 12

• Add code-execution class after AutoPenBench gate finding

  failed:
  ci#11:Commit75b3804pushedbyZaidux
  main

  Oct 9, 20264s
  More actions for workflow run 11

• Run phase-7 gate against AutoPenBench; find RCE taxonomy gap

  failed:
  ci#10:Commita618a7fpushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 10

• Add scorer evaluation, applicability-aware baseline, phase 6 results

  failed:
  ci#9:Commit941eca1pushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 9

• Add applicability function, rewire task suite, document data spec

  failed:
  ci#8:Commit2bc934bpushedbyZaidux
  main

  Oct 9, 20264s
  More actions for workflow run 8

• Add paired A/B harness, synthetic task suite, reasoning-truncation guard

  failed:
  ci#7:Commit24cfa41pushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 7

• Fix oracle provider: zen endpoint with space-bunny-free

  failed:
  ci#6:Commit69857f8pushedbyZaidux
  main

  Oct 9, 20264s
  More actions for workflow run 6

• Add strategy blind-spot detection, ungrounded-claim checks, oracle ba…

  failed:
  ci#5:Commit5a2ff6cpushedbyZaidux
  main

  Oct 9, 20263s
  More actions for workflow run 5