A security device watches over the secure functionality in a computer system. This “watcher” security device may be integrated within the computer system or may be separate from it. The security device queries the secure functionality to determine whether the state of the secure functionality is acceptable. If no satisfactory state exists, or if no response is received, then a signal is transmitted. The signal may be auditory (a buzzer) or visual (a flashing light) in order to signal to any user that the secure functionality has been compromised. Optionally, human input devices may be disabled, or a monitoring service notified, in conjunction with or in lieu of the signal. If the secure functionality includes a secret shared between the secure functionality and the user, then the security device may signal the secret. For example, where the secret is visual, the security device may display the secret. Where there is more than one element of secure functionality in the computer system, the security device may separately watch and report on more than one element of secure functionality. The security device may also display status information regarding the computer system. Some or all of the security device may be distributed via a trusted distribution infrastructure.