firmware-release-security-adjudication

Task gallery / Security / cyber / Version 1

Reference solution (oracle): Pass
No-op resistance check: Pass
QA verdict: snorkel review: needs-revision 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-36c36db89046cb22 · 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 / effort	Verifier result [95% Wilson]	Diagnostic labels / valid	HARNESS_ERROR · excluded	Outcome classes

Legacy evidence

Model / effort	Verifier result [95% Wilson]	Diagnostic labels / valid	HARNESS_ERROR · excluded	Outcome classes
Claude Opus 4.8 · default (unverified)	2/5 · 40.0% [11.8%, 76.9%]	0 / 5	0	GOOD_FAILURE: 0 BAD_FAILURE: 0 GOOD_SUCCESS: 0 BAD_SUCCESS: 0
GPT-5.5 · default (unverified)	4/5 · 80.0% [37.6%, 96.4%]	0 / 5	0	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 / model	Date / source	Reward / eligibility	Diagnostic class / source	Resources	Evidence
`tbench-task__8oThqBn` oracle@@	Not recorded repo	1 · valid	No diagnostic label derived	Tokens: — Turns: — Cost: —	Not publicly available
`tbench-task__LQAw9MH` oracle@@	Not recorded repo	1 · valid	No diagnostic label derived	Tokens: — Turns: — Cost: —	Not publicly available
`tbench-task__rMW6Dhv` nop@@	Not recorded repo	0 · valid	No diagnostic label derived	Tokens: — Turns: — Cost: —	Not publicly available
`tbench-task__a7yRc2y` oracle@@	Not recorded repo	1 · valid	No diagnostic label derived	Tokens: — Turns: — Cost: —	Not publicly available
`tbench-task__RQoTDh5` Claude Opus 4.8 · default (unverified)	Not recorded repo	0 · valid	No diagnostic label derived	Tokens: 36,292 Turns: — Cost: $3.1434	Not publicly available
`tbench-task__2aNT62P` Claude Opus 4.8 · default (unverified)	Not recorded repo	0 · valid	No diagnostic label derived	Tokens: 39,987 Turns: — Cost: $3.3188	Not publicly available
`tbench-task__YoeDUuz` Claude Opus 4.8 · default (unverified)	Not recorded repo	1 · valid	No diagnostic label derived	Tokens: 40,551 Turns: — Cost: $3.2332	Not publicly available
`tbench-task__BEw8SyL` Claude Opus 4.8 · default (unverified)	Not recorded repo	1 · valid	No diagnostic label derived	Tokens: 52,033 Turns: — Cost: $4.6327	Not publicly available
`tbench-task__rSwNR6S` Claude Opus 4.8 · default (unverified)	Not recorded repo	0 · valid	No diagnostic label derived	Tokens: 36,841 Turns: — Cost: $2.7564	Not publicly available
`tbench-task__54bHjSu` GPT-5.5 · default (unverified)	Not recorded repo	1 · valid	No diagnostic label derived	Tokens: 17,919 Turns: — Cost: $3.0625	Not publicly available
`tbench-task__CVWTfDM` GPT-5.5 · default (unverified)	Not recorded repo	1 · valid	No diagnostic label derived	Tokens: 16,861 Turns: — Cost: $2.1944	Not publicly available
`tbench-task__CdGPJNA` GPT-5.5 · default (unverified)	Not recorded repo	1 · valid	No diagnostic label derived	Tokens: 14,711 Turns: — Cost: $2.1125	Not publicly available
`tbench-task__3oj7nmE` GPT-5.5 · default (unverified)	Not recorded repo	1 · valid	No diagnostic label derived	Tokens: 20,539 Turns: — Cost: $3.0129	Not publicly available
`tbench-task__B6AtspS` GPT-5.5 · default (unverified)	Not recorded repo	0 · valid	No diagnostic label derived	Tokens: 13,368 Turns: — Cost: $1.8572	Not publicly available