outpost-arena · Syncvals

Task gallery / Games / Version 1

Reference solution (oracle): Pass
No-op resistance check: Pass
QA verdict: Not recorded
Aggregate eligibility: Eligible version

The task

You're entered as the programmer in a bot tournament for Outpost, a small supply-line wargame. Two sides fight over a 16 by 16 map: you claim ground by walking on it, depots pay a trickle of supply, and any unit that loses its trail of friendly ground back to headquarters starts starving. Take the enemy HQ and the game ends right there; otherwise whoever holds more ground when the clock runs out wins, and a dead-eve…

Excerpt only; not a complete executable task specification. Task ID task_4868dff5988142f9 · version 1. Selection status: not fully recorded.

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
Claude Opus 5 · default (unverified)	4/5 · 80.0% [37.6%, 96.4%]	5 / 5	0	GOOD_FAILURE: 1 BAD_FAILURE: 0 GOOD_SUCCESS: 4 BAD_SUCCESS: 0

Legacy evidence

Model / effort	Verifier result [95% Wilson]	Diagnostic labels / valid	HARNESS_ERROR · excluded	Outcome classes

Difficulty, with uncertainty

Selected-best difficulty: 20.0% [3.6%, 62.4%]. This is one minus the best measured current-row pass rate among rows with at least five valid runs.

The maximum is selected from the observed rows, so the estimate is optimistic. Its Wilson interval is not adjusted for model selection. “Hard” requires qualifying GOOD_FAILURE evidence; the threshold flag additionally requires the interval and recorded quality gates to meet the documented rule. This is not independent validation. Threshold flag: no.

Every recorded run

7 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
`trial_327e9f28b03d41c3` nop@@	Oct 10, 2026 console	0 · valid	No diagnostic label derived	Tokens: — Turns: — Cost: —	Not publicly available
`trial_7aa7b9a098a944c2` oracle@@	Oct 10, 2026 console	1 · valid	No diagnostic label derived	Tokens: — Turns: — Cost: —	Not publicly available
`trial_183312e2b40a47f0` Claude Opus 5 · default (unverified)	Oct 10, 2026 console	1 · valid	GOOD_SUCCESS frozen-classifier	Tokens: 123,475 Turns: 139 Cost: $17.0285	Not publicly available
`trial_2392b392b5b64f08` Claude Opus 5 · default (unverified)	Oct 10, 2026 console	1 · valid	GOOD_SUCCESS frozen-classifier	Tokens: 151,461 Turns: 185 Cost: $24.8613	Not publicly available
`trial_3b5f8e2ffac248a6` Claude Opus 5 · default (unverified)	Oct 10, 2026 console	1 · valid	GOOD_SUCCESS frozen-classifier	Tokens: — Turns: — Cost: —	Not publicly available
`trial_a0b59ad86eaa4ad0` Claude Opus 5 · default (unverified)	Oct 10, 2026 console	0 · valid	GOOD_FAILURE frozen-classifier	Tokens: 136,059 Turns: 112 Cost: $15.5059	Not publicly available
`trial_b539fa6238ee4efd` Claude Opus 5 · default (unverified)	Oct 10, 2026 console	1 · valid	GOOD_SUCCESS frozen-classifier	Tokens: — Turns: — Cost: —	Not publicly available