Task gallery / Security / cyber / Version 1
Task evidence / repo

bell-vault-custody-gate

legacy-onlyChosen benchmark version
Reference solution (oracle)
Pass
No-op resistance check
Pass
QA verdict
snorkel review: accept x1
Aggregate eligibility
Eligible version

The task

An instruction excerpt is not available in the public snapshot.

Excerpt only; not a complete executable task specification. Task ID review-709c83cb10d627bf · version 1. Selection status: not fully recorded.

repo ↗

Results by model and effort

This exact task version only. Every rate is passes / valid runs with a Wilson 95% interval; n≤2 remains counts only. Macro aggregation does not apply within a single task. Baselines are summarized in the quality gates, not ranked. A passed no-op resistance check means that doing nothing did not solve the task.

Current evidence

Model / effortVerifier result [95% Wilson]Diagnostic labels / validHARNESS_ERROR · excludedOutcome classes

Legacy evidence

Model / effortVerifier result [95% Wilson]Diagnostic labels / validHARNESS_ERROR · excludedOutcome classes
Claude Opus 4.8 · default (unverified)0/5 · 0.0% [0.0%, 43.4%]0 / 50
GOOD_FAILURE: 0
BAD_FAILURE: 0
GOOD_SUCCESS: 0
BAD_SUCCESS: 0
GPT-5.5 · default (unverified)3/5 · 60.0% [23.1%, 88.2%]0 / 50
GOOD_FAILURE: 0
BAD_FAILURE: 0
GOOD_SUCCESS: 0
BAD_SUCCESS: 0

Every recorded run

14 recorded trials, including controls. Diagnostic labels do not replace verifier reward. Labels and quality checks are source declarations, not independently attested here. Internal evidence remains private; a missing public URL is shown explicitly. Records with invalid or absent rewards are not scored.

Run / modelDate / sourceReward / eligibilityDiagnostic class / sourceResourcesEvidence
tbench-task__2fAj8Z5
oracle@@		Not recorded
repo		1 · validNo diagnostic label
derived		Tokens: —
Turns: —
Cost: —		Not publicly available
tbench-task__sgSZtwU
oracle@@		Not recorded
repo		1 · validNo diagnostic label
derived		Tokens: —
Turns: —
Cost: —		Not publicly available
tbench-task__3LEYMMR
nop@@		Not recorded
repo		0 · validNo diagnostic label
derived		Tokens: —
Turns: —
Cost: —		Not publicly available
tbench-task__GWF5Rje
oracle@@		Not recorded
repo		1 · validNo diagnostic label
derived		Tokens: —
Turns: —
Cost: —		Not publicly available
tbench-task__MHfaKSs
Claude Opus 4.8 · default (unverified)		Not recorded
repo		0 · validNo diagnostic label
derived		Tokens: 27,188
Turns: —
Cost: $2.2853		Not publicly available
tbench-task__p78dk2W
GPT-5.5 · default (unverified)		Not recorded
repo		1 · validNo diagnostic label
derived		Tokens: 21,749
Turns: —
Cost: $3.6032		Not publicly available
tbench-task__QBPyAKq
GPT-5.5 · default (unverified)		Not recorded
repo		1 · validNo diagnostic label
derived		Tokens: 24,682
Turns: —
Cost: $4.8202		Not publicly available
tbench-task__MPfracn
Claude Opus 4.8 · default (unverified)		Not recorded
repo		0 · validNo diagnostic label
derived		Tokens: 25,566
Turns: —
Cost: $2.0017		Not publicly available
tbench-task__77QD97D
GPT-5.5 · default (unverified)		Not recorded
repo		1 · validNo diagnostic label
derived		Tokens: 18,741
Turns: —
Cost: $4.0190		Not publicly available
tbench-task__DWQgtmL
Claude Opus 4.8 · default (unverified)		Not recorded
repo		0 · validNo diagnostic label
derived		Tokens: 26,321
Turns: —
Cost: $1.9283		Not publicly available
tbench-task__Zq7t3ox
GPT-5.5 · default (unverified)		Not recorded
repo		0 · validNo diagnostic label
derived		Tokens: 22,902
Turns: —
Cost: $4.1855		Not publicly available
tbench-task__fAyEeyv
Claude Opus 4.8 · default (unverified)		Not recorded
repo		0 · validNo diagnostic label
derived		Tokens: 23,668
Turns: —
Cost: $1.9545		Not publicly available
tbench-task__a8sqBM3
GPT-5.5 · default (unverified)		Not recorded
repo		0 · validNo diagnostic label
derived		Tokens: 12,199
Turns: —
Cost: $3.7567		Not publicly available
tbench-task__RSiSdxu
Claude Opus 4.8 · default (unverified)		Not recorded
repo		0 · validNo diagnostic label
derived		Tokens: 19,359
Turns: —
Cost: $1.4763		Not publicly available