/

CVE-2026-50085

Severity:

HIGH RISK

(8.6)

Missing Authentication (CWE-306) in Aqara MQTT Service

CVE-2026-50085

Analyzed

Vulnerability type

Category:

Missing Authentication for Critical Function

CWE:

CWE-306

Impact:

Remote Device Control and Potential Takeover

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

The Aqara Board service (op-test.aqara.com) forwards arbitrary MQTT command payloads to the platform's HiveMQ broker without enforcing authentication on the caller.

An attacker on the internet can send crafted MQTT messages that may control or manipulate connected devices, and when chained with related flaws, this behavior can contribute to full remote takeover of affected devices.

Affected context

This vulnerability occurs in the cloud-facing Aqara Board service that proxies MQTT traffic to a backend HiveMQ broker.

It is relevant when the service:

  • Is reachable over the internet (e.g., op-test.aqara.com),
  • Accepts MQTT command payloads from unauthenticated clients, and
  • Forwards those commands to backend infrastructure that influences device behavior or device configuration.

The risk increases when devices trust commands from this broker by default and do not require additional verification.

Exploit preconditions

  • Attack surface: Public network exposure of the Aqara Board service endpoint (op-test.aqara.com).
  • Required access: None; an attacker only needs the ability to initiate network connections to the service and speak the MQTT or corresponding API protocol.
  • Trigger point: Submission of arbitrary MQTT command payloads that the service forwards to the HiveMQ broker as if they originated from a trusted source.
  • Why exploitation is possible: The service does not enforce authentication or robust authorization checks before accepting and relaying MQTT commands, allowing untrusted clients to access critical messaging functionality.

Why this happens

The underlying issue is a missing authentication and authorization layer around a critical control function: the ability to send MQTT commands into the platform's broker.

Instead of requiring a device key, user credential, or token-based identity, the service treats inbound messages as implicitly trusted and forwards them to HiveMQ.

This aligns with CWE-306: Missing Authentication for Critical Function, where an operation that can significantly affect system state is exposed without verifying the caller's identity or permissions.

Potential impact

  • An attacker may issue arbitrary MQTT commands affecting connected devices.
  • Device configuration or firmware settings could be manipulated remotely.
  • Cloud-side state related to devices (e.g., pairing, scenes, or automation rules) could be altered.
  • When combined with related vulnerabilities (CVE-2026-50082, CVE-2026-50083, CVE-2026-50084), an attacker could remotely and unauthenticatedly take over affected devices.
  • Service reliability and availability may be degraded through malicious or high-volume MQTT message floods.

General mitigation guidance

Effective mitigation generally focuses on enforcing strict identity and access controls on critical messaging pathways:

  • Require strong authentication (e.g., mutual TLS, token-based credentials, or signed MQTT clients) for any endpoint that can issue control-related MQTT commands.
  • Introduce fine-grained authorization checks that validate whether a given client is allowed to act on specific topics or devices.
  • Adopt a least-privilege model for MQTT topics, separating diagnostic, telemetry, and control channels with distinct access policies.
  • Harden broker and proxy configurations so that anonymous or unauthenticated traffic cannot reach control topics by default.
  • Implement defense-in-depth at the device layer: devices can verify command provenance (e.g., signatures or per-device keys) rather than trusting broker-delivered commands implicitly.

How teams detect issues like this

Detection usually combines architectural review with targeted monitoring:

  • Perform code and configuration reviews focusing on network-exposed services that forward or broker MQTT commands, verifying that authentication and authorization are present for all critical operations.
  • Use static analysis and configuration scanning to flag anonymous or unauthenticated connections permitted on brokers or proxies, especially for control topics.
  • Inspect access logs for MQTT gateways and brokers for patterns of unauthenticated or anomalous command traffic, such as unknown client identifiers targeting many devices.
  • Enforce security policies in CI/CD that prevent deploying services with open, unauthenticated access to privileged MQTT operations.

Sources

SourceLink
github.comgithub.com/xn0tsa/theres-no-place-like-home
runzero.comrunzero.com/advisories/aqara-board-iot-insecure-debug-api-cve-2026-50085

Frequently asked

questions

Yes. If the Aqara Board service is reachable from the internet, an attacker can send MQTT command payloads without prior access or authentication, subject to network routing and any external controls.

No local credentials are required for the vulnerable service itself. However, the actual impact depends on how the backend broker and devices interpret and trust the forwarded commands.

Introduce mandatory authentication and authorization checks on all endpoints that can publish or forward MQTT commands, ensuring only authenticated and authorized clients can access control topics or device-specific operations.

Treat all control-plane messaging interfaces as security-critical, design them with strong identity requirements, enforce topic-level access control, and include authentication checks as a non-negotiable requirement during design reviews and pre-deployment security testing.

Find out what's already
exploitable in your codebase.