CVE-2025-32018
HighSummary
Podatność w aplikacji Cursor, wersje 0.45.0 do 0.48.6, pozwalała na nieautoryzowane modyfikacje plików poza otwartym obszarem roboczym w wyniku działania agenta Cursor. Wymagało to jednak celowego wywołania przez użytkownika lub złośliwie przygotowany kontekst, co czyniło skuteczne wykorzystanie tej podatności mało prawdopodobnym w praktyce.
Risk Assessment
Ryzyko dla organizacji polega na możliwości nieautoryzowanej modyfikacji plików, co może prowadzić do utraty danych lub wprowadzenia niepożądanych zmian. Mimo że wykorzystanie tej podatności jest mało prawdopodobne, należy być świadomym potencjalnych zagrożeń.
Recommendation
Zaleca się aktualizację aplikacji Cursor do wersji 0.48.7, aby usunąć tę podatność. Dodatkowo warto monitorować działania użytkowników oraz wprowadzić odpowiednie procedury bezpieczeństwa.
Related vulnerabilities
- CVE-2026-102240Critical
A vulnerability was found in Netcore NAP930 0.1.241010.141410 in the eval function of the /www/cgi-bin/network_tools file of the Network Tools CGI component. Manipulation of the sid argument results in OS command injection. The attack can be performed remotely and the exploit has been made public.
- CVE-2026-101354Critical
A security flaw was discovered in FAST FAC1203R 20200116_2.0.4 in the _tWlanTask function of the MmtAtePrase Parser component. Manipulation results in a stack-based buffer overflow. The attacker must have access to the local network, and the exploit has been released.
- CVE-2026-102361Critical
mall4j through 4.0 contains a missing authentication vulnerability in the PUT /user/updatePwd endpoint that allows unauthenticated attackers to reset any storefront account password. Attackers can supply a target username in the request body to overwrite passwords without verification, enabling account takeover and access to orders and personal data.
- CVE-2026-101264Critical
A vulnerability was determined in Ziroom ZHOME A0101 1.0.1.0 in an unknown function of the /api/ZRnetwork/set_passwd file. Manipulation of the password1 argument causes command injection. The attack can be initiated remotely and the exploit has been publicly disclosed.
- CVE-2026-101263Critical
A vulnerability was found in Ziroom ZHOME A0101 1.0.1.0 in some unknown processing of the /api/ZRQos/set_online_client file. Manipulation of the mac argument results in command injection. The attack can be launched remotely and the exploit has been made public.
- CVE-2026-101262Critical
A vulnerability in Ziroom ZHOME A0101 version 1.0.1.0 affects unknown code in the file /api/ZRQos/set_online_client. Manipulation of the argument ip leads to command injection. The attack can be performed remotely, and the exploit is publicly available.
- CVE-2026-101261Critical
A flaw in Ziroom ZHOME A0101 version 1.0.1.0 affects an unknown part of the file /api/ZRnetwork/firstSetup_wifi. Manipulation of the argument login_pwd leads to command injection. The attack can be performed remotely, and the exploit is published.
- CVE-2026-101260Critical
A vulnerability was detected in Ziroom ZHOME A0101 version 1.0.1.0, affecting unknown functionality of the file /api/ZRnetwork/firstLogin. Manipulation of the argument firstLogin results in command injection. The attack can be performed remotely, and the exploit is public.
- CVE-2026-102268Critical
PyJWT is a Python implementation of JSON Web Token standards. Prior to version 2.14.0, is_pem_format in jwt/utils.py does not recognize every PEM representation accepted by the cryptography loader. When an application mixes HMAC and asymmetric algorithms and supplies a mutated public-key PEM as raw key bytes, HMACAlgorithm.prepare_key treats the unrecognized asymmetric public key as an HMAC secret. An attacker who knows the public key can forge authenticated HMAC tokens. This issue is fixed in version 2.14.0.
- CVE-2026-101187Critical
A weakness in Ziroom ZHOME A0101 version 1.0.1.0 affects the function pop_usb_device in the file usr/lib/lua/luci/controller/api/zrUsb.lua of the USB Device Management API component. Manipulation of the argument path causes command injection. The attack can be performed remotely, and the exploit is publicly available.
Original NVD description (English source)
Cursor is a code editor built for programming with AI. In versions 0.45.0 through 0.48.6, the Cursor app introduced a regression affecting the set of file paths the Cursor Agent is permitted to modify automatically. Under specific conditions, the agent could be prompted, either directly by the user or via maliciously crafted context, to automatically write to files outside of the opened workspace. This behavior required deliberate prompting, making successful exploitation highly impractical in real-world scenarios. Furthermore, the edited file was still displayed in the UI as usual for user review, making it unlikely for the edit to go unnoticed by the user. This vulnerability is fixed in 0.48.7.
Vulnerability data from NVD (NIST) · CISA KEV · EPSS

