/

CVE-2026-72581

Severity:

HIGH RISK

(8.6)

SSRF (CWE-918) in XiaoAI Patch Authentication API Endpoint

CVE-2026-72581

Deferred

Vulnerability type

Category:

Server-Side Request Forgery

CWE:

CWE-918

Impact:

Internal Network Access and Service Discovery

Severity:

HIGH RISK

(8.6)

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 server-side request forgery vulnerability in the XiaoAI Patch authentication API allows a remote attacker to influence where the Xiaomi smart speaker sends HTTP requests. When the endpoint accepts an unvalidated destination URL, the attacker may reach internal services, scan network resources, or request external URLs through the device.

Affected context

The vulnerability occurs in deployments using the affected authentication endpoint and its URL-based redirection flow. Risk is higher when the speaker can reach private network services, administrative interfaces, or cloud metadata endpoints and when the destination is derived directly from a user-supplied request parameter.

Exploit preconditions

  • Attack surface: The /auth HTTP endpoint exposed by the application
  • Required access: Network access sufficient to submit requests to the endpoint; authentication requirements depend on deployment configuration
  • Trigger point: Processing of the user-controlled url POST parameter
  • Why exploitation is possible: The endpoint uses the supplied destination to initiate or redirect HTTP requests without adequately validating the target or restricting access to private network ranges

Why this happens

The underlying bug is an unchecked server-side HTTP request destination. User-controlled URL data is treated as a trusted routing instruction rather than as untrusted input requiring validation and policy enforcement.

This maps to CWE-918: Server-Side Request Forgery, where an application makes requests to attacker-influenced locations from a trusted server or device. The insecure pattern can expose the device's network position and connectivity to resources that are not directly reachable by the attacker.

Potential impact

  • An attacker may enumerate reachable internal hosts and services
  • An attacker could access internal HTTP endpoints or administrative interfaces
  • Sensitive responses may be disclosed through the application or device
  • The vulnerability can increase the risk of compromise of adjacent network services
  • External services may receive requests originating from the affected device

General mitigation guidance

Mitigation generally focuses on reducing trust in user-supplied URLs and enforcing an explicit outbound request policy.

Common defensive patterns include:

  • Allowing requests only to an explicit, minimal set of approved destinations
  • Validating URL schemes, hostnames, ports, and resolved addresses before making requests
  • Blocking loopback, link-local, private, multicast, and other restricted address ranges
  • Rechecking destinations after DNS resolution and preventing unsafe redirect chains
  • Separating authentication redirects from server-side fetching where possible
  • Applying outbound network controls and least-privilege service permissions
  • Avoiding exposure of response bodies or detailed connection errors to untrusted users

How teams detect issues like this

Organizations can look for SSRF risks through:

  • Static analysis rules identifying HTTP clients whose destinations derive from request parameters
  • Code review checks for missing URL allowlists, private-address filtering, and redirect restrictions
  • Runtime logging of outbound requests, resolved addresses, ports, and redirect targets
  • Network monitoring for unexpected connections from the speaker or application host to internal ranges
  • Policy enforcement that limits egress to approved services and protocols
  • Automated security review tools that flag user-controlled URL construction and server-side request flows

Sources

SourceLink
github.comgithub.com/duhow/xiaoai-patch
github.comgithub.com/duhow/xiaoai-patch/blob/main/api/main.py

Frequently asked

questions

Yes. A remote attacker may be able to trigger the vulnerable request flow through the exposed authentication endpoint, subject to the endpoint's deployment-specific access controls.

The attacker needs network access to submit a request to the affected endpoint. The exact authentication requirement depends on how the XiaoAI Patch service is deployed.

Do not use arbitrary user-provided URLs for server-side requests. Prefer fixed destinations or a narrowly scoped allowlist, and validate resolved addresses and redirect targets.

Combine strict URL validation with outbound firewall policies, private-address blocking, careful DNS handling, least-privilege permissions, and monitoring of unexpected egress traffic.

Find out what's already
exploitable in your codebase.