CVE-2025-64763
Envoyproxy Envoy ≤ 1.33.13
Raw vector
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:N/A:NSummary
CVE-2025-64763 is a low-severity Protection Mechanism Failure (CWE-693) vulnerability in Envoyproxy Envoy. Its CVSS base score is 3.7 (Low).
Operationally, exploitation aligns with the MITRE ATT&CK technique Exploit Public-Facing Application (T1190); ranked at the 21th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.
The strongest mitigations our analysis identified map to SC-7 (Boundary Protection) and SI-10 (Information Input Validation) — see the control section below for these in your framework.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2025-201099
Vulnerability Data
Envoy is a high-performance edge/middle/service proxy. In 1.33.12, 1.34.10, 1.35.6, 1.36.2, and earlier, when Envoy is configured in TCP proxy mode to handle CONNECT requests, it accepts client data before issuing a 2xx response and forwards that data to the…
more
upstream TCP connection. If a forwarding proxy upstream from Envoy then responds with a non-2xx status, this can cause a de-synchronized CONNECT tunnel state. By default Envoy continues to allow early CONNECT data to avoid disrupting existing deployments. The envoy.reloadable_features.reject_early_connect_data runtime flag can be set to reject CONNECT requests that send data before a 2xx response when intermediaries upstream from Envoy may reject establishment of a CONNECT tunnel.
- CWE(s)
Related Threats
MITRE ATT&CK Enterprise TechniquesAI
Why these techniques?
The vulnerability allows an attacker to send data before a 2xx response in CONNECT requests, potentially leading to a desynchronized tunnel state. This can be exploited against a publicly accessible Envoy proxy (T1190: Exploit Public-Facing Application) to cause unexpected behavior or state confusion in the proxy's handling of CONNECT tunnels.
CVEs Like This One
Affected Assets
Mitigating Controls
Control response
Mitigating Controls (NIST 800-53 r5) AI
Envoy functions as a boundary proxy; SC-7 requires boundary devices to enforce correct CONNECT tunnel establishment before forwarding client data, directly blocking the early-data forwarding flaw.
Requires validation of CONNECT request sequencing so that payload data is rejected until a 2xx response confirms tunnel setup, eliminating the desynchronization vector.
Enforces information flow rules that data may traverse the TCP proxy only after successful CONNECT negotiation, preventing premature upstream forwarding.
Mitigating Controls (NIST CSF 2.0) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.
Enforcing authentication directly implements a core protection mechanism whose absence or misuse is the CWE.
Defining and enforcing access authorizations is a protection mechanism; proper use prevents the CWE.
Cryptographic and integrity controls are protection mechanisms whose correct deployment mitigates the CWE.
Encryption and integrity protections for transit are explicit protection mechanisms.
Logical network protections are protection mechanisms whose failure matches the CWE.
Hardened configuration baselines ensure protection mechanisms are correctly applied and maintained.
Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.
Systematic verification that security mechanisms operate according to defined standards reduces the likelihood that protection mechanisms are bypassed or disabled.
Hardening devices, disabling vulnerable protocols, and maintaining accurate network diagrams reduce the likelihood that a protection mechanism is misconfigured or left in a weak state.
Requiring defined escalation paths, crisis activation criteria, and coordination procedures strengthens the overall protection mechanism so that a single control failure is less likely to leave the organization exposed.
By requiring a documented categorization and decision process for security events, the control ensures that protection mechanisms are not bypassed or ignored when anomalies occur.
Identifying and remediating control weaknesses that contributed to an incident reduces the likelihood that protection mechanisms will fail again.
Post-incident analysis that feeds updated risk assessments and additional controls directly reduces the chance that previously exploited weaknesses will recur.