CVE-2026-26205 in opa-envoy-plugin
Summary
by MITRE • 02/19/2026
opa-envoy-plugun is a plugin to enforce OPA policies with Envoy. Versions prior to 1.13.2-envoy-2 have a vulnerability in how the `input.parsed_path` field is constructed. HTTP request paths are treated as full URIs when parsed; interpreting leading path segments prefixed with double slashes (`//`) as authority components, and therefore dropping them from the parsed path. This creates a path interpretation mismatch between authorization policies and backend servers, enabling attackers to bypass access controls by crafting requests where the authorization filter evaluates a different path than the one ultimately served. Version 1.13.2-envoy-2 fixes the issue.
If you want to get best quality of vulnerability data, you may have to visit VulDB.
Analysis
by VulDB Data Team • 02/19/2026
The vulnerability identified as CVE-2026-26205 affects the opa-envoy-plugin, a critical component that enforces Open Policy Agent policies within Envoy proxy environments. This plugin serves as a bridge between policy enforcement and network traffic handling, making it a prime target for attackers seeking to circumvent security controls. The flaw manifests in the plugin's handling of HTTP request paths, specifically in how the input.parsed_path field is constructed during request processing. The issue stems from improper URI parsing logic that fails to maintain consistency between policy evaluation and actual request handling, creating a fundamental mismatch that can be exploited by malicious actors.
The technical root cause of this vulnerability lies in the plugin's interpretation of HTTP request paths containing double slash prefixes. When processing requests with paths beginning with double slashes, the system treats these leading path segments as authority components rather than path elements, effectively stripping them from the parsed path representation. This behavior creates a divergence between what the authorization policies evaluate and what the backend servers actually process, as the policy engine operates on the modified path while the actual service receives the original full URI. The vulnerability specifically impacts versions prior to 1.13.2-envoy-2, indicating that the developers have acknowledged and addressed this path interpretation discrepancy in their subsequent releases.
From an operational impact perspective, this vulnerability enables attackers to bypass access controls by crafting malicious requests that exploit the path interpretation mismatch. An attacker could construct a request where the authorization filter evaluates a path that grants access, while the backend server processes a different path that may be restricted or sensitive. This creates a scenario where legitimate access control mechanisms fail to prevent unauthorized access, potentially allowing privilege escalation or data exposure. The vulnerability aligns with CWE-20, "Improper Input Validation," and represents a path traversal issue that can be leveraged for privilege escalation attacks, particularly in environments where the plugin enforces access controls for sensitive resources.
The security implications extend beyond simple access control bypass, as this vulnerability can be exploited in conjunction with other attack vectors to create more sophisticated breaches. Attackers can craft requests that appear to target authorized paths while actually accessing restricted resources, making detection more challenging. This type of vulnerability is particularly dangerous in microservices architectures where Envoy proxies mediate communication between services and OPA policies enforce security boundaries. The issue can be categorized under ATT&CK technique T1078.004, "Valid Accounts: Cloud Accounts," as it may enable attackers to gain access to resources that would otherwise be protected by proper authorization controls. Organizations using this plugin should implement immediate mitigations including upgrading to version 1.13.2-envoy-2 or later, and conducting thorough security reviews of their authorization policies to identify potential exploitation vectors.