A critical (CVSS 10.0) header injection vulnerability in the Axios HTTP client library, CVE-2026-40175, can be chained with a proxy bypass (CVE-2025-62718) to enable Server-Side Request Forgery (SSRF) and cloud metadata exfiltration.
CONFIRMED: No active exploitation observed. The vulnerability is not listed on CISA's KEV. The EPSS score is low (0.403%), indicating a low probability of exploitation in the next 30 days despite the critical CVSS score.
Immediate action is to identify all server-side applications utilizing the Axios library and prepare for patching. Monitor for vendor advisories and patch releases.
Intelligence Metadata
First Reported Bymsrc.microsoft.com
Outbreak Velocity0 minutes to 1 unique domains
Consensus1 source across 1 unique domain
CVSS Score10.0
EPSS Score0.00403 (4.03rd percentile)
CISA KEV StatusNot Listed
Exploit Risk Score0.4016
Race LabelFRESH EXPLOIT
Technical Analysis
The vulnerability, identified as CVE-2026-40175, is a header injection flaw within the Axios HTTP client library. According to the initial report from Microsoft, this can lead to unrestricted cloud metadata exfiltration. This flaw appears to be part of an attack chain involving CVE-2025-62718, which is described as a `NO_PROXY` hostname normalization bypass. The inferred attack path begins with an attacker leveraging the normalization bypass to circumvent proxy configurations, resulting in a Server-Side Request Forgery (SSRF) condition. Subsequently, the attacker can use the header injection vulnerability to craft malicious requests to internal network resources, with a primary focus on cloud instance metadata services (e.g., AWS IMDS, Azure IMDS). Successful exploitation would allow an unauthenticated, remote attacker to exfiltrate sensitive data, including temporary cloud credentials, from the underlying server infrastructure.
Affected Systems
Vendor: Axios
Affected Versions:
Axios (specific vulnerable versions not provided in source intelligence)
Exposure: Server-side applications (e.g., Node.js) that use the Axios library to make HTTP requests where request parameters, such as the URL or headers, can be influenced by untrusted external input.
Blast Radius: High. Axios is a ubiquitous JavaScript HTTP client library. A significant number of server-side applications using a vulnerable version for outbound requests could be potential ingress points.
Intelligence Context
The threat picture for CVE-2026-40175 presents a significant conflict between severity and probability. The CVSS 10.0 score reflects the maximum theoretical impact, likely due to the risk of complete information disclosure via cloud credential theft without authentication. However, this is sharply contrasted by the low EPSS score of 0.403%, which predicts a very low likelihood of exploitation in the wild within the next 30 days. The intelligence is currently single-sourced from Microsoft (msrc.microsoft.com) and is not yet listed in CISA's KEV, corroborating the absence of observed exploitation campaigns. This profile suggests a vulnerability that, while technically severe, may require specific, non-default configurations or complex preconditions to exploit, reducing its immediate attractiveness to attackers. For defenders, this warrants prioritizing asset discovery and monitoring over emergency, out-of-band patching.
Remediation & Defense
Investigate exposed assets. Identify all server-side codebases utilizing the Axios library. Monitor for an official vendor patch and apply it based on standard risk assessment.
Patch Status: Not available
Patch Version: Not available
Workarounds:
Implement strict, allow-list-based validation on all user-supplied input used to construct URLs or headers in server-side Axios requests.
Utilize network egress filtering to block or log all requests from application servers to known cloud metadata service IP addresses (e.g., 169.254.169.254).
Detection Hints:
Monitor for outbound HTTP requests from application servers to unexpected internal destinations, particularly cloud metadata endpoints.
Inspect application logs for evidence of malformed URLs or unusual headers in requests initiated by the Axios library.