Whitaker Domains is a registrar. You work on the support desk, which uses the internal side of the control panel: domain records, transfer logs, abuse reports. The billing ledger is finance's, and you have never been able to open it.
Everything else under /internal answers you normally. The ledger answers with a page that has Server: nginx on it and no trace of the panel, which is a Node application that has never rendered an HTML error in its life.
One change to this request triggers the vulnerability. The challenge is to use your security intuition to work out what to change. Find the value the server shouldn’t trust, change it, and send it. The response tells you whether you were right.
Hint. Read the internal billing ledger.
“A deny rule for an endpoint covers every request the application will route to that endpoint.”
Authorise the ledger in the application, against the actor, where the route is already known. If the web server is genuinely the control, deny the prefix rather than the exact URI, and add a test that asks for the protected path with a trailing slash and expects the same refusal. The general rule is that the component doing the matching should be the component doing the routing.
Anywhere an exact string is the guard and a lenient router is the target: nginx location =, an ingress rule with an exact path type, a WAF rule pinned to one URI, an allowlist of endpoints in a gateway. Trailing slashes, a repeated slash, a trailing dot and a case change are all one keystroke each, and a framework that treats them as the same route will happily serve what the guard did not recognise.
One Request Away is a daily series of challenges. They’re like code review, but for HTTP requests: in each one you’re one request away from triggering a vulnerability, and your goal is to identify what you need to change.