openssl_encrypt CVE-2026-81691

Exploitation requires the attacker to hold a network-path position between the vulnerable application and its keyserver - achievable via ARP spoofing on a shared LAN segment, a rogue Wi-Fi access point, a compromised network device, or ISP-level traffic interception. The vulnerable code paths are specifically the login and register_with_email functions; other library functions are not described as affected. The CVSS 4.0 vector (PR:N/UI:N) confirms no authentication or user interaction is required beyond achieving network positioning. Exploitation is constrained to deployments where the library is configured with http:// keyserver URLs or where the host is unconfigured - installations that independently enforce HTTPS-only URLs at the application configuration layer have a reduced practical attack surface, though the underlying defect remains unpatched.

The CVSS 4.0 score of 8.7 (AV:N/AC:L/AT:N/PR:N/UI:N/VC:H) accurately reflects theoretical severity: no privileges or user interaction are required, and credential exposure is complete. However, practical exploitability is tempered by the prerequisite of network path positioning - the attacker must intercept traffic between the application and its keyserver, requiring ARP poisoning, a rogue access point, ISP-level access, or a compromised network device. No EPSS data was supplied in the source intelligence, leaving exploitation probability relative to the broader ecosystem unquantified. KEV status is negative and no public exploit code exists, suggesting no active opportunistic exploitation campaigns. The credential types exposed - passwords, JWTs, and client_ids - enable full account takeover, making this a high-consequence flaw despite the network positioning prerequisite. Real-world risk is elevated for deployments communicating with keyservers over untrusted networks (cloud-to-cloud over public internet, mobile clients, shared hosting).

An attacker who has gained a network-path position between the application server and its keyserver - for example via ARP poisoning on a shared LAN or by operating a rogue access point - passively captures HTTP traffic. When any user triggers the login or register_with_email function, their client_id, password, and JWT are transmitted as cleartext HTTP, readable with standard capture tools such as tcpdump or Wireshark. The attacker replays these credentials directly against the keyserver to achieve full account takeover without any further exploitation steps.