Ask HN: Where does this adversarial prize mechanism break?

I'm looking for adversarial review of a mechanism design problem involving real money. There is a fixed on-chain prize pool and a physical cache at a pre-committed deterministic location. Clues unlock for a fee, and each unlock adds to the prize pool. Anyone can attempt a claim at any time. No rule changes after deployment. The operator commits the location in advance, which introduces a trust assumption. Claims are validated against committed coordinates within a tolerance radius. First round is active with a small initial pool. Contract: https://etherscan.io/address/0x04d71b7bCD29BA68D7Ec92d41f444814EcB42119 Spec: https://hashclue.com/technical-documents Site: https://hashclue.com Questions I am keen to answer: - What are the economic attack vectors? (griefing, collusion, info asymmetry) - Is the clue pricing floor structurally unstable at small pool sizes? - Does the operator trust assumption make this fundamentally flawed? - Is this effectively just a lottery under a different interface? - What happens when rational players model each other's clue-buying behavior? - Is there a dominant strategy that collapses the game before it plays out? - Would you participate? Why or why not? I'm more interested in structural criticism than product feedback.