CVE-2023-28322

An attacker can exploit a vulnerability in curl to send malicious commands during TELNET communication, potentially allowing them to execute arbitrary code on the system. This can happen if an application using curl accepts user input without properly checking it, making it particularly risky for applications that rely on user-provided data.

This vulnerability allows an attacker to potentially intercept sensitive information during data transfers because the curl tool may fail to upgrade certain HTTP requests to secure HTTPS when multiple requests are made at the same time. This issue occurs only when using curl versions prior to 7.88.0 and can lead to unprotected data being sent over the internet instead of the intended secure connection.

An attacker can trick curl into using an insecure HTTP connection instead of the intended secure HTTPS by manipulating the URL with special characters that confuse the software's security checks. This vulnerability occurs when the URL contains IDN characters that are converted to ASCII, allowing the attacker to bypass the HSTS protection that should enforce secure connections.

This vulnerability allows an attacker to manipulate data sent in a POST request by exploiting a flaw in how the curl library handles reused connections, potentially leading to the wrong data being sent or causing the application to crash. It occurs when a connection that was previously used for a PUT request is reused for a POST request without properly resetting the data handling, which could happen in applications that frequently switch between these types of requests.

This vulnerability allows an attacker to trick curl into using an insecure HTTP connection instead of the intended secure HTTPS connection by manipulating the URL with special characters. This can happen when the URL includes international domain names that get converted to ASCII, making it possible for the attacker to bypass security checks designed to enforce HTTPS.