dssrf CVE-2026-44232

Exploitation requires that the vulnerable application (1) uses dssrf versions prior to 1.3.0 with the is_url_safe() function for URL validation, (2) accepts attacker-controlled input that becomes part of HTTP request URLs (such as webhook callbacks, proxy targets, link expansion services, or image fetch URLs), (3) runs in an environment where IPv6 networking is functional or where the Node.js HTTP client processes IPv6 bracket notation (default behavior in Node.js http/https modules), and (4) has network connectivity to internal resources or metadata services that would be blocked by proper SSRF protections. Exploitation is NOT limited to environments with native IPv6 routing - IPv4-mapped IPv6 addresses (::ffff:x.x.x.x) work even in IPv4-only networks because they are processed at the application layer before network transmission. No authentication or user interaction is required; the attack succeeds through server-side request handling. The vulnerability is most dangerous in cloud deployments (AWS, GCP, Azure) where metadata services at 169.254.169.254 are reachable, but also enables access to any internal HTTP services on private IPs, Docker APIs, Kubernetes endpoints, or localhost services when crafted as IPv6 loopback [::1].

Without CVSS vector or EPSS data, risk assessment relies on technical impact and exploit accessibility. The vulnerability presents HIGH real-world risk for applications using dssrf < 1.3.0 as an SSRF defense layer, because publicly available exploit code exists demonstrating 12 distinct IPv6 bypass techniques against the library's core security function. Attack complexity is LOW - exploitation requires only crafting URLs with IPv6 bracket notation, no special privileges, and the POC script provides ready-to-use payloads. Technical impact is CRITICAL when the protected application runs in cloud environments (AWS EC2, GCP Compute, Azure VMs) where IMDS endpoints at [::ffff:169.254.169.254] expose IAM credentials, user-data scripts, and instance metadata. Impact extends to internal service enumeration, Docker daemon access (if bound to IPv6), Kubernetes API servers, and private network pivoting. The vulnerability is particularly dangerous because it creates a false sense of security - developers believing dssrf protects against SSRF will not implement defense-in-depth controls. However, exploitation is NOT fully automatic across all deployments: it requires that (1) the vulnerable application actually makes HTTP requests to attacker-controlled URLs, (2) the underlying Node.js HTTP client supports IPv6 (most do by default), and (3) IPv6 networking is functional in the deployment environment (increasingly common but not universal). Organizations using dssrf 1.0.2-1.2.x in cloud environments or IPv6-enabled networks should treat this as an urgent patching priority. No CISA KEV status indicates this is not yet exploited at scale, but the comprehensive POC and cloud metadata targeting suggest rapid weaponization risk.

An attacker identifies a Node.js web application using dssrf for URL validation before fetching user-specified resources (common in proxy services, webhook handlers, image processors, or link preview features). The attacker submits a POST request with URL parameter 'http://[::ffff:169.254.169.254]/latest/meta-data/iam/security-credentials/' which dssrf incorrectly marks as safe. The application fetches this URL, reaches the AWS EC2 instance metadata service via IPv6-mapped IPv4 address, and returns IAM temporary credentials to the attacker, enabling full compromise of the cloud workload's AWS permissions.