CodiMD through 2.5.4 has a CSP-based protection mechanism against XSS through uploaded SVG documents containing JavaScript, but it can be bypassed in certain cases of different-origin file storage, such as AWS S3. NOTE: it can be considered a user error if AWS is employed for hosting untrusted JavaScript content, but the selected architecture within AWS does not have components that are able to insert Content-Security-Policy headers.

CodiMD - Missing Image Access Controls and Unauthorized Image Access

CodiMD allows realtime collaborative markdown notes on all platforms. CodiMD before 2.5.4 is missing authentication and access control vulnerability allowing an unauthenticated attacker to gain unauthorised access to image data uploaded to CodiMD. CodiMD does not require valid authentication to access uploaded images or to upload new image data. An attacker who can determine an uploaded image's URL can gain unauthorised access to uploaded image data. Due to the insecure random filename generation in the underlying Formidable library, an attacker can determine the filenames for previously uploaded images and the likelihood of this issue being exploited is increased. This vulnerability is fixed in 2.5.4.

CodiMD through 2.2.0 has a CSP-based protection mechanism against XSS through uploaded JavaScript content, but it can be bypassed by uploading a .html file that references an uploaded .js file.

HackMD CodiMD <2.5.2 is vulnerable to Denial of Service.

Cross-site Scripting in Hackmd.io Notes lead by HTML Injection

CodiMD allows realtime collaborative markdown notes on all platforms. The notebook feature of Hackmd.io permits the rendering of iframe `HTML` tags with an improperly sanitized `name` attribute. This vulnerability enables attackers to perform cross-site scripting (XSS) attacks via DOM clobbering. This vulnerability is fixed in 2.5.4.

CodiMD 1.3.1, when Safari is used, allows XSS via an IFRAME element with allow-top-navigation in the sandbox attribute, in conjunction with a data: URL.