KaratumTestnet · pilotOpen Explorer

Verdict

Task bugfix-cg-business-days-02 · Claude Code · Opus (external, code)

◐ Partial Evidence E3

failed: hidden_tests_pass

Agent claim (E0)

The agent said success: “Fixed businessDaysBetween: loop bound was Math.min(x,y) instead of Math.max — the loop never iterated, always returning 0. Now iterates from min to max and negates when a>b. All 12 tests pass.”

A claim is self-reported and never decides the verdict. Here the chain state contradicts the claim.

Outcome Spec checks

  • ✓ Passchanges_made
    Expected
    ≥1 changed file
    Actual
    src/business-days.js M +4 -2
  • ✓ Passonly_allowed_paths_changed
    Expected
    only src/business-days.js, src/csv.js, src/invoice.js
    Actual
    in scope
  • ✓ Passno_test_harness_tampering
    Expected
    none
    Actual
    none
  • ✓ Passvisible_tests_pass
    Expected
    all passed, 0 failed
    Actual
    12 passed, 0 failed
  • ✗ Failhidden_tests_pass
    Expected
    all passed, 0 failed
    Actual
    46 passed, 1 failed

Task spec

Type software.bugfix

{
  "id": "bugfix-cg-business-days-02",
  "type": "software.bugfix",
  "params": {
    "fixture": "cg-business-days-02",
    "allowedPaths": [
      "src/business-days.js",
      "src/csv.js",
      "src/invoice.js"
    ]
  },
  "version": 1
}

On-chain evidence

Transactions
None
Evidence class
E3
Verdict hash
0x17614e0dc7f11966d65e5be0804a43cca2fb637f673b2b1100c568223125153e
Attestation
View attestation on EAS ↗
Attestation tx
0x4b5835d443fba2683336a459eb888240112c250bf3b7b50fe0d598492bffefab

Run

Steps
8
Gas used
0
LLM cost
$0.309
Latency
47.0 s
Finished