Skip to content

Trust

Trust and security

Hypr Markets is software for running a terminal on Hyperliquid. It never holds client funds. This page says plainly what is in place today and what is not done yet.

Who runs the platform

The company that operates the platform and issues its invoices.
The platform's own policies and the templates terminals start from.

How we protect accounts and data

A summary of the practices in place. Details that would help an attacker are left out.
  • No custody of client funds

    Clients trade from their own wallets on Hyperliquid and sign every order themselves. Neither the platform nor the terminal can move a client's money.

  • Wallet sign-in with optional two-factor codes

    Operators sign in by signing a message with their wallet. They can add authenticator codes, which are then asked again before sensitive changes. Platform administrators must use them.

  • Sessions and requests

    Session tokens are random and stored only as a hash. The session cookie is not readable by page scripts, and every change carries a separate anti-forgery token.

  • Sites are kept apart

    Every request is checked against the terminal it belongs to. Automated tests try to read and change other terminals' data, and must fail.

  • Fee routing is tested

    Which address receives the trading fee is decided by one rule, shared by every service, with tests for each plan and billing state.

  • Less data, kept for less time

    Logs remove IP addresses, wallet addresses and emails. Visit counts store no IP or browser fingerprint. Old sessions and sign-in codes are deleted on a schedule.

  • No spending keys on servers

    Subscription payments go to a treasury wallet whose keys are kept off every server; our security policy requires it to be a multisig. The servers only watch for incoming payments.

  • Limits on abuse

    Requests are rate limited, uploads are size-limited and images are re-encoded, and tenant sites carry a strict content security policy.

Testing and reviews

What has been checked, by whom, and what is still open.
Internal security review
Completed on Nov 10, 2026: code review, invariant tests and fuzzing across the api, tenant sites and payments.
Red-team exercise
Completed on Nov 11, 2026 by a tester who did not build the platform: no high-severity findings, two medium findings fixed, three low findings documented.
Independent penetration test
Loading

Service status

Live state of the parts of the platform your clients use.

System status. We do not publish uptime history yet.

Report a security issue

We read every report and reply to the address you write from.

A dedicated security contact is not published yet. Until it is, use the support address in the help center and mark the message as a security report.