
Cloud Vulnerability DB
A community-led vulnerabilities database
CVE-2026-0672 is a CRLF injection vulnerability in Python's http.cookies.Morsel class that allows user-controlled cookie names, values, and parameters to inject arbitrary HTTP headers into messages. It affects CPython versions prior to 3.15.0 (including actively maintained branches 3.10 through 3.14) and was disclosed on January 20, 2026, by the Python Software Foundation (PSF). The vulnerability carries a CVSS v4.0 base score of 6.0 (Medium) (Feedly, EUVD).
The root cause is improper neutralization of CRLF sequences (CWE-93) in the http.cookies.Morsel class within Python's standard library. Prior to the fix, the Morsel.__setitem__(), Morsel.setdefault(), Morsel.set(), and SimpleCookie.output() methods did not validate cookie names, values, or parameters for C0 control characters (Unicode range \x00-\x1F and \x7F), including carriage return (\r) and line feed (\n). An attacker with the ability to influence cookie values passed to these methods — for example, via user-supplied input in a web application — could embed CRLF sequences to inject additional HTTP headers into server responses. The fix adds a _has_control_character() helper using the regex [\x00-\x1F\x7F] and raises a CookieError if any control character is detected (CPython PR #143920, CPython commit 95746b3).
Successful exploitation enables HTTP response splitting and header injection attacks, which can lead to session hijacking, cache poisoning, cross-site scripting via injected headers, and web log tampering. The integrity impact is rated High (attacker can manipulate HTTP response structure), while confidentiality impact is Low (limited information disclosure). Availability is not directly affected. The vulnerability is network-exploitable and requires low privileges, but attack requirements (user-controlled input reaching http.cookies.Morsel) must be present (Feedly, EUVD).
http.cookies.Morsel (or SimpleCookie) that accepts user-controlled input for cookie names, values, or parameters without prior sanitization.legitimate_value\r\nInjected-Header: malicious-content.Morsel.set(), Morsel.__setitem__(), or Morsel.setdefault().SimpleCookie.output() or similar, embedding the injected CRLF sequence into the HTTP response headers.Set-Cookie headers with embedded newlines, or injected headers like Location, Content-Type, or custom headers not set by the application).%0d%0a, %0D%0A, \r\n) in cookie-related parameters or POST body fields.CookieError: Control characters are not allowed in cookies exceptions (on patched versions), which may indicate attempted exploitation.The Python Software Foundation released patches for all actively maintained branches: CPython 3.10, 3.11, 3.12, 3.13, and 3.14 (backported via PRs #144089–#144094), with the fix merged to main on January 20, 2026 (commit 95746b3). Downstream distributions including Debian, Ubuntu (USN-8018-1, USN-8018-2, USN-8018-3), SUSE, Fedora, Amazon Linux 2, Amazon Linux 2023, Red Hat Enterprise Linux, and AlmaLinux have released updated packages. Users should upgrade to the latest patched Python release for their branch. As a workaround, applications should sanitize all user-supplied input before passing it to http.cookies.Morsel, explicitly rejecting any strings containing control characters (\x00-\x1F, \x7F) (CPython PR #143920, Debian LTS, Ubuntu USN-8018-1).
The fix was authored by Seth Michael Larson (PSF Security Developer-in-Residence) and reviewed by multiple CPython core developers, reflecting a well-coordinated disclosure and patch process. The Python security announcement was published to the security-announce@python.org mailing list and the oss-security list. Downstream Linux distributions responded promptly with security advisories and package updates across Debian, Ubuntu, SUSE, Fedora, Red Hat, and Amazon Linux. No significant controversy or notable social media discussion beyond routine security update announcements was observed (Python Security Announce, oss-sec).
Source: This report was generated using AI
Free Vulnerability Assessment
Evaluate your cloud security practices across 9 security domains to benchmark your risk level and identify gaps in your defenses.
Get a personalized demo
"Best User Experience I have ever seen, provides full visibility to cloud workloads."
"Wiz provides a single pane of glass to see what is going on in our cloud environments."
"We know that if Wiz identifies something as critical, it actually is."