Spring AMQP CVE-2026-41701

Exploitation requires the application to be using RabbitTemplate.sendAndReceive() with an explicitly configured fixed reply queue - this is a non-default deployment pattern; applications using the default temporary per-request queue model are not vulnerable. The CVSS vector confirms high privileges are required (PR:H), meaning the attacker must already hold elevated access to the AMQP broker environment or internal network segment. Attack complexity is rated High (AC:H), indicating a race condition or precise timing element must be satisfied to inject or intercept the reply before the legitimate response arrives. No user interaction is required (UI:N). These conditions collectively limit exploitation to insider threats, compromised privileged service accounts, or attackers with significant existing foothold in the messaging infrastructure. Standard external attackers without broker-level access cannot exploit this vulnerability.

The CVSS 4.4 Medium score is calibrated by significant mitigating factors: Attack Complexity is rated High (AC:H), meaning specific race or positioning conditions must be met, and Privileges Required is High (PR:H), constraining exploitation to already-privileged broker or application actors. The network attack vector (AV:N) widens the theoretical exposure surface, and the changed scope (S:C) signals that successful exploitation can affect components beyond the vulnerable RabbitTemplate instance - for example, other consumers sharing the same fixed reply queue. Confidentiality and integrity impacts are both Low (C:L/I:L), consistent with partial message interception or injection rather than full system compromise. No EPSS score was provided in source data; one cannot be inferred. There is no CISA KEV listing and no publicly known exploit code, suggesting this is a theoretical or targeted risk rather than a commodity threat. The real-world risk is further constrained by the non-default deployment requirement of a fixed reply queue. Organizations that rely on the default temporary-queue pattern are not affected. The combination of PR:H and AC:H makes this a low-priority item for most environments, though the S:C signal warrants attention in multi-tenant or high-throughput broker deployments where reply queue isolation is critical.

An attacker who has already obtained high-privilege access to the RabbitMQ broker - for example, a compromised internal service account or a malicious insider - observes an in-flight sendAndReceive() correlation ID (e.g., '42') and trivially predicts subsequent values. The attacker races to publish a crafted reply message with correlation ID '43' to the fixed reply queue before the legitimate server reply arrives, causing the RabbitTemplate consumer to accept the forged reply as authentic, potentially injecting false data into the application's business logic.