The Twitter subsystem in Apple iOS before 7 does not require API conformity for access to Twitter daemon interfaces, which allows attackers to post Tweets via a crafted app that sends direct requests to the daemon.

The Social subsystem in Apple iOS before 7 does not properly restrict access to the cache of Twitter icons, which allows physically proximate attackers to obtain sensitive information about recent Twitter interaction via unspecified vectors.

WebKit in Apple iOS before 7 allows remote attackers to bypass the Same Origin Policy and obtain potentially sensitive information about use of the window.webkitRequestAnimationFrame API via an IFRAME element.

Passcode Lock in Apple iOS before 7.0.2 on iPhone devices allows physically proximate attackers to bypass an intended passcode requirement, and dial arbitrary telephone numbers, by making a series of taps of the emergency-call button to trigger a NULL pointer dereference.

Passcode Lock in Apple iOS before 7.0.2 does not properly manage the lock state, which allows physically proximate attackers to bypass an intended passcode requirement, and open the Camera app or read the list of all recently opened apps, by leveraging unspecified transition errors.

Passcode Lock in Apple iOS before 7.0.3 on iPhone devices allows physically proximate attackers to bypass the passcode-failure disabled state by leveraging certain incorrect visibility of the passcode-entry view after use of the Phone app.

socketfilterfw in Application Firewall in Apple Mac OS X before 10.9 does not properly implement the --blockApp option, which allows remote attackers to bypass intended access restrictions via a network connection to an application for which blocking was configured.

CoreGraphics in Apple Mac OS X before 10.9, when display-sleep mode is used, does not ensure that screen locking blocks the visibility of all windows, which allows physically proximate attackers to obtain sensitive information by reading the screen.

CoreGraphics in Apple Mac OS X before 10.9 allows local users to bypass secure input mode and log an arbitrary application's keystrokes via a hotkey event registration.

LaunchServices in Apple Mac OS X before 10.9 does not properly restrict Unicode characters in filenames, which allows context-dependent attackers to spoof file extensions via a crafted character sequence.

App Sandbox in Apple Mac OS X before 10.9 allows attackers to bypass intended sandbox restrictions via a crafted app that uses the LaunchServices interface to specify process arguments.

Power Management in Apple Mac OS X before 10.9 does not properly handle the interaction between locking and power assertions, which allows physically proximate attackers to obtain sensitive information by reading a screen that should have transitioned into the locked state.

The Screen Lock implementation in Apple Mac OS X before 10.9 does not immediately accept Keychain Status menu Lock Screen commands, and instead incorrectly relies on a certain timeout setting, which allows physically proximate attackers to obtain sensitive information by reading a screen that should have transitioned into the locked state.

The Screen Lock implementation in Apple Mac OS X before 10.9, when hibernation and autologin are enabled, does not require a password for a transition out of hibernation, which allows physically proximate attackers to obtain access by visiting an unattended workstation in the hibernating state.

Apple Mac OS X before 10.9 does not preserve a certain administrative system-preferences setting across software updates, which allows context-dependent attackers to bypass intended access restrictions in opportunistic circumstances by leveraging an unintended security configuration after the completion of an update.

Smart Card Services in Apple Mac OS X before 10.9 does not properly implement certificate-revocation checks, which allows remote attackers to cause a denial of service (Smart Card usage outage) by interfering with the revocation-check procedure.

The syslog implementation in Apple Mac OS X before 10.9 allows local users to obtain sensitive information by leveraging access to the Guest account and reading console-log messages from previous Guest sessions.

Apple Safari before 6.1.1 and 7.x before 7.0.1 allows remote attackers to bypass the Same Origin Policy and discover credentials by triggering autofill of subframe form fields.

Adobe ColdFusion 10 before Update 12 allows remote attackers to read arbitrary files via unspecified vectors.

Secunia CSI Agent 6.0.0.15017 and earlier, 6.0.1.1007 and earlier, and 7.0.0.21 and earlier, when running on Red Hat Linux, uses world-readable and world-writable permissions for /etc/csia_config.xml, which allows local users to change CSI Agent configuration by modifying this file.