/

CVE-2026-51152

Severity:

CRITICAL RISK

(9.1)

SSRF (CWE-918) in QD Unauthenticated HAR Test API Service

CVE-2026-51152

Deferred

Vulnerability type

Category:

Server-Side Request Forgery

CWE:

CWE-918

Impact:

Unauthorized access to internal services and cloud metadata, with possible credential disclosure and further network compromise

Severity:

CRITICAL RISK

(9.1)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

A critical server-side request forgery vulnerability in QD's unauthenticated /har/test endpoint allows remote attackers to make the server issue HTTP requests to attacker-selected destinations. Because the endpoint does not validate URL schemes, hosts, or IP ranges and disables TLS certificate verification, internal services and cloud metadata endpoints may be reachable.

Affected context

The vulnerability occurs in QD deployments where the /har/test endpoint is exposed and accepts user-supplied JSON that is passed to Fetcher.build_request().

Risk is higher when the QD server can reach private network services, loopback addresses, container control planes, or cloud metadata endpoints. The absence of authentication broadens exposure to unauthenticated remote clients.

Exploit preconditions

  • Attack surface: The HTTP /har/test endpoint
  • Required access: Network access to the endpoint; authentication is not required
  • Trigger point: Processing attacker-controlled JSON that defines an outbound HTTP request
  • Why exploitation is possible: The request builder does not adequately constrain URL schemes, destinations, hostnames, or IP ranges, allowing the server to act as a proxy toward otherwise inaccessible resources

Why this happens

The underlying bug is server-side request forgery caused by treating client-controlled request destinations as trusted.

Fetcher.build_request() constructs an HTTP request from user-supplied JSON without enforcing an allowlist of schemes and destinations or blocking private, loopback, link-local, and reserved address ranges. Disabling TLS certificate validation further weakens transport trust. This behavior maps to CWE-918: Server-Side Request Forgery, because an attacker can influence a server into making requests on their behalf.

Potential impact

  • An attacker may access internal administrative or service endpoints
  • Cloud instance metadata and temporary credentials could be disclosed
  • Internal network topology and service information could be exposed
  • The QD server could be used to reach systems that are not externally accessible
  • Follow-on compromise or unauthorized actions could occur if sensitive services trust the QD server
  • Disabling certificate verification could allow interception or acceptance of untrusted TLS endpoints

General mitigation guidance

Mitigation generally focuses on restricting outbound request destinations and reducing trust in client-provided network parameters.

Common defensive patterns include:

  • Allowing only explicitly required URL schemes, hosts, ports, and destination networks
  • Resolving hostnames and validating every resulting IP address before connecting
  • Blocking loopback, private, link-local, multicast, reserved, and cloud metadata ranges
  • Rechecking destinations after redirects and preventing DNS rebinding
  • Preserving TLS certificate and hostname verification by default
  • Requiring authentication and authorization for request-proxying functionality
  • Applying network egress controls and least-privilege service permissions
  • Returning minimal response details to reduce internal information disclosure

How teams detect issues like this

Organizations can look for SSRF risks through:

  • Static analysis rules identifying HTTP clients whose URLs derive from request data
  • Code review signals such as unrestricted URL fields, disabled TLS verification, and missing redirect validation
  • Tests covering private, loopback, link-local, reserved, and metadata IP ranges
  • Network monitoring for unexpected QD connections to internal services or metadata addresses
  • Egress firewall policies that alert on or block unauthorized destination ranges
  • API policy checks requiring authentication and destination allowlists for server-side fetch features

Automated security review tools may help identify user-controlled URLs that reach outbound request libraries.

Sources

SourceLink
gist.github.comgist.github.com/kurokoleung/98e950e3e05205fc57d2029ad68a702e
qd-today.github.ioqd-today.github.io/qd/

Frequently asked

questions

Yes. An attacker who can reach the unauthenticated /har/test endpoint may submit requests remotely without possessing an account.

Network connectivity to the exposed endpoint is generally sufficient. The attacker does not need direct access to the QD host or the targeted internal resources.

Restrict outbound requests with strict scheme, host, port, and IP-range allowlists; validate destinations after DNS resolution and redirects; and retain normal TLS certificate verification.

Require authentication for proxy-like endpoints, enforce egress filtering, treat all request destinations as untrusted input, and add automated tests and code-review checks for SSRF patterns.

Find out what's already
exploitable in your codebase.