Signal : Security Vulnerabilities, CVEs,

Signal Messenger for Android 4.24.8 may expose private information when using "disappearing messages." If a user uses the photo feature available in the "attach file" menu, then Signal will leave the picture in its own cache directory, which is available to any application on the system.

The Open Whisper Signal app before 2.23.2 for iOS allows physically proximate attackers to bypass the screen locker feature via certain rapid sequences of actions that include app opening, clicking on cancel, and using the home button.

js/views/message_view.js in Open Whisper Signal (aka Signal-Desktop) before 1.10.1 allows XSS via a URL.

Open Whisper Signal (aka Signal-Desktop) through 1.10.1 allows XSS via a resource location specified in an attribute of a SCRIPT, IFRAME, or IMG element, leading to JavaScript execution after a reply, a different vulnerability than CVE-2018-10994. The attacker needs to send HTML code directly as a message, and then reply to that message to trigger this vulnerability. The Signal-Desktop software fails to sanitize specific HTML elements that can be used to inject HTML code into remote chat windows when replying to an HTML message. Specifically the IMG and IFRAME elements can be used to include remote or local resources. For example, the use of an IFRAME element enables full code execution, allowing an attacker to download/upload files, information, etc. The SCRIPT element was also found to be injectable. On the Windows operating system, the CSP fails to prevent remote inclusion of resources via the SMB protocol. In this case, remote execution of JavaScript can be achieved by referencing the script on an SMB share within an IFRAME element, for example: <IFRAME src=\\DESKTOP-XXXXX\Temp\test.html> and then replying to it. The included JavaScript code is then executed automatically, without any interaction needed from the user. The vulnerability can be triggered in the Signal-Desktop client by sending a specially crafted message and then replying to it with any text or content in the reply (it doesn't matter).

Open Whisper Signal (aka Signal-Desktop) before 1.15.0-beta.10 allows information leakage.

The image rendering component (createGenericPreview) of the Open Whisper Signal app through 2.29.0 for iOS fails to check for unreasonably large images before manipulating received images. This allows for a large image sent to a user to exhaust all available memory when the image is displayed, resulting in a forced restart of the device.

Open Whisper Signal (aka Signal-Desktop) through 1.23.1 and the Signal Private Messenger application through 4.35.3 for Android are vulnerable to an IDN homograph attack when displaying messages containing URLs. This occurs because the application produces a clickable link even if (for example) Latin and Cyrillic characters exist in the same domain name, and the available font has an identical representation of characters from different alphabets.

The Signal Private Messenger application before 4.47.7 for Android allows a caller to force a call to be answered, without callee user interaction, via a connect message. The existence of the call is noticeable to the callee; however, the audio channel may be open before the callee can block eavesdropping.

The WebRTC component in the Signal Private Messenger application through 4.47.7 for Android processes videoconferencing RTP packets before a callee chooses to answer a call, which might make it easier for remote attackers to cause a denial of service or possibly have unspecified other impact via malformed packets. NOTE: the vendor plans to continue this behavior for performance reasons unless a WebRTC design change occurs

Signal Desktop before 1.29.1 on Windows allows local users to gain privileges by creating a Trojan horse %SYSTEMDRIVE%\node_modules\.bin\wmic.exe file.

Signal Private Messenger Android v4.59.0 and up and iOS v3.8.1.5 and up allows a remote non-contact to ring a victim's Signal phone and disclose currently used DNS server due to ICE Candidate handling before call is answered or declined.

The Signal app before 5.34 for iOS allows URI spoofing via RTLO injection. It incorrectly renders RTLO encoded URLs beginning with a non-breaking space, when there is a hash character in the URL. This technique allows a remote unauthenticated attacker to send legitimate looking links, appearing to be any website URL, by abusing the non-http/non-https automatic rendering of URLs. An attacker can spoof, for example, example.com, and masquerade any URL with a malicious destination. An attacker requires a subdomain such as gepj, txt, fdp, or xcod, which would appear backwards as jpeg, txt, pdf, and docx respectively.

Signal Desktop before 6.2.0 on Windows, Linux, and macOS allows an attacker to modify conversation attachments within the attachments.noindex directory. Client mechanisms fail to validate modifications of existing cached files, resulting in an attacker's ability to insert malicious code into pre-existing attachments or replace them completely. A threat actor can forward the existing attachment in the corresponding conversation to external groups, and the name and size of the file will not change, allowing the malware to masquerade as another file. NOTE: the vendor disputes the relevance of this finding because the product is not intended to protect against adversaries with this degree of local access.

Signal Desktop before 6.2.0 on Windows, Linux, and macOS allows an attacker to obtain potentially sensitive attachments sent in messages from the attachments.noindex directory. Cached attachments are not effectively cleared. In some cases, even after a self-initiated file deletion, an attacker can still recover the file if it was previously replied to in a conversation. (Local filesystem access is needed by the attacker.) NOTE: the vendor disputes the relevance of this finding because the product is not intended to protect against adversaries with this degree of local access.