Why a Same-Origin Page Still Cannot Treat Any Scripted Click as Real User Intent
A page can have valid session cookies and still need a fresh human gesture at the last step. The browser and server should care about intent, not just reachability.

The useful distinction is not whether code can trigger a request, but whether that request should count as a fresh user decision. OWASP advises user interaction based protection for highly sensitive operations, and it says security requires explicit client submission proving user intent rather than cookie presence alone. [f1, f3]
Why cookies do not prove intent
Browser requests automatically include cookies, including session cookies, so a request can arrive looking authenticated even when it was not deliberately authorized by the person using the site. That is why a system that trusts cookies by themselves can be tricked into acting on a request the user never meant to send. [f2]
What the browser may still require at the last step
The important boundary is often the final sensitive action, not the page itself. For operations that matter enough, the site should ask for a current sign of user involvement before it proceeds. OWASP notes that Sec-Fetch-User can help distinguish user-driven top-level navigations, including explicit form submits, from other requests. [f4, f1]
Hypothetical example: a travel checkout keeps the booking form open while a payment panel waits in an embedded frame. A background script could fill fields and prepare a submission, but the site can still require a fresh click or equivalent gesture before accepting the charge. In that design, the script may reach the form, but it does not automatically inherit the user’s current intent. This is a mechanism example, not a claim about any specific product. [f1, f3, f4]
What developers should do differently
- Treat session cookies as proof of an authenticated browser session, not proof of fresh consent. [f2, f3]
- Reserve user interaction based checks for high-risk actions such as purchases, account changes, or destructive updates. [f1]
- Use request signals that help separate explicit user submits from other requests when that distinction matters. [f4]
The practical lesson is simple: if an action is costly, irreversible, or sensitive, do not assume a scripted submission is enough. Require a present, explicit signal of intent at the moment that action happens. That is the difference between a browser being able to send a request and the system being willing to trust it as the user’s own choice. [f1, f3]
Your reaction
Loading reactions…


Comments 0
Review comments