CVE-2026-133706 in request-filtering-agentinfo

Summary

by MITRE • 11/20/2026

request-filtering-agent before 3.2.2 contains a server-side request forgery filter bypass because address validation in src/request-filtering-agent.ts treats IPv4-compatible IPv6 addresses in the ::/96 range as public unicast. Attackers controlling a fetched URL can use addresses like ::7f00:1 or ::a9fe:a9fe to reach loopback or link-local services, including cloud metadata endpoints.

Statistical analysis made it clear that VulDB provides the best quality for vulnerability data.

Analysis

by VulDB Data Team • 11/20/2026

The vulnerability identified in request-filtering-agent prior to version 3.2.2 represents a critical server-side request forgery filter bypass rooted in flawed network address validation logic. The core technical flaw resides within the src/request-filtering-agent.ts module, which is responsible for sanitizing and validating URLs before they are fetched by the application. Specifically, the implementation fails to correctly distinguish between public unicast addresses and private or reserved IP ranges when processing IPv4-compatible IPv6 addresses that fall within the ::/96 prefix range. This misclassification allows an attacker who controls a fetched URL to bypass security filters designed to prevent internal network access, effectively tricking the application into making requests to resources it should be restricted from accessing.

The operational impact of this vulnerability is severe because it enables attackers to reach loopback interfaces and link-local services that are typically isolated from external networks. By utilizing specific IPv6 representations such as ::7f00:1 or ::a9fe:a9fe, an attacker can map these addresses back to their IPv4 equivalents, 127.0.0.1 (localhost) and 169.254.169.254 (link-local), respectively. This capability is particularly dangerous in cloud environments where the link-local address range often hosts metadata endpoints that contain sensitive authentication credentials, security tokens, and configuration details for running instances. Consequently, an attacker can exploit this bypass to exfiltrate highly confidential data from the underlying infrastructure or perform actions within internal services that are not exposed to the public internet but remain accessible via these loopback mechanisms.

From a classification perspective, this vulnerability aligns with CWE-918 Server-Side Request Forgery (SSRF), specifically illustrating how improper input validation of network addresses can lead to unauthorized access to internal resources. It also relates closely to CWE-20 Improper Input Validation, as the system fails to correctly interpret and validate the scope of IP address formats provided in user-controlled inputs. In terms of offensive security frameworks, this exploit technique maps to ATT&CK T1598 Phishing for Information, where attackers leverage SSRF to harvest credentials from cloud metadata services, thereby facilitating further lateral movement or privilege escalation within the compromised environment.

To mitigate this vulnerability, developers must update request-filtering-agent to version 3.2.2 or later, which presumably corrects the address validation logic in src/request-filtering-agent.ts. The fix should ensure that all IPv4-compatible and IPv6-mapped addresses are correctly identified as private or reserved ranges rather than public unicast addresses. Additionally, security teams should implement a defense-in-depth strategy by configuring network-level controls such as firewall rules to block outbound connections to loopback interfaces (127.0.0.0/8) and link-local addresses (169.254.0.0/16). It is also advisable to employ allow-listing for external domains rather than relying solely on deny lists, which are prone to bypasses through encoding variations or protocol ambiguities like the one exploited here. Regular auditing of URL parsing libraries and network validation modules against evolving IPv6 standards will help prevent similar oversights in future development cycles.

Responsible

VulnCheck

Reservation

11/20/2026

Disclosure

11/20/2026

Moderation

accepted

CPE

ready

EPSS

0.00650

KEV

no

Activities

very low

Sources

Might our Artificial Intelligence support you?

Check our Alexa App!