Trust, before you pay.

No to all three. There is no endpoint that accepts your brokerage credentials: we never hold a key, never move a dollar, and never place the order. Your agent asks whether an action is inside the limits you set, we answer allow, ask a human or no, and the account stays exactly where it was. Account state arrives with the question and is never stored. The decision log keeps a hash of it instead, so a refusal stays auditable and your positions stay yours.

No, because the limits are not a prompt: they are checked outside the model, on every path that moves money rather than only on orders. A recurring buy, an account setting, money wired out, a cancelled protective position: that last one is where naive guardrails leak, because an agent blocked from buying can simply schedule the buy instead. A refusal cannot be overwritten, and attempts to overwrite one are counted rather than quietly rejected.

Nothing of ours can break a trade, because we are not in the execution path. When our own package cannot get a verdict it refuses to act rather than guessing, which is the safe direction to fail in. You do not have to take our word for whether the engine is up: this endpoint runs the real engine on request and answers 503 if it is not enforcing, and a monitor checks it every hour.

Live, on brokerage trading and card spending. Here is the part most products would hide: the public counters read zero, because the engine is enforcing but no live agent has produced a decision yet, and we would rather show a zero you can verify on the public endpoint than a number you cannot reproduce. Two honest limits: on a card we can refuse a charge before it happens but we cannot stop an agent that already holds the card number, and prediction markets are designed and deliberately not built.