SSH, Actually 100% signal, 0% noise.

One-Time Codes

concept
depth 3/3 − deeper + surface

An OTP (One-Time Password) is a short-lived numeric code generated by an app or device. You type it after your password, so a stolen password alone is no longer enough.

How it works: a shared secret exists between you and the service; time or counters generate changing codes from it; you type the current code after your password.

The delivery mechanism matters more than people expect:

  • SMS codes—a short numeric code texted to a phone number. Weak: SMS is vulnerable to SIM-swapping, carrier attacks, interception, and social engineering. The phone number—not the physical device—is the security anchor, which is a fundamental design flaw.
  • Authenticator apps (TOTP)—an app generates a rotating code from a shared secret and the current time. Better: codes are short-lived and never transmitted over the network. Still flawed: they are phishable and replayable—an attacker who tricks you into entering the code can immediately use it.
  • Enterprise tokens—centrally managed hardware or software tokens with short validity windows, logging, and rapid revocation. Stronger operationally, but still code-based, so real-time phishing remains possible.

Where one-time codes sit in the larger landscape: they are layered defense, not a structural fix. The two limits are the same limits passwords have, just softened:

  • Codes can still be phished. Anything you type, you can be tricked into typing on the wrong site—and a live attacker relays it in real time.
  • Secrets still exist on servers. TOTP works because the server holds the same shared secret your app does. The server can still lose it.

That is why every code-based method—email link, SMS, TOTP, enterprise tokens—sits below the cryptographic tiers on the MFA strength ladder. FIDO2 and its descendants don't rotate the shared secret faster; they remove it. The server verifies a signature instead, and there is nothing to type, capture, or replay.