A Study of Data Store-based Home Automation

Kaushal Kafle, Kevin Moran, Sunil Manandhar, Adwait Nadkarni, Denys Poshyvanyk

This study, presented at CODASPY 2019, provides a systematic security evaluation of Data Store-Based (DSB) home automation platforms, specifically focusing on Google’s Nest and Philips Hue. While these platforms offer seamless automation through "routines," the researchers uncover significant architectural flaws that allow low-integrity devices to indirectly control high-security ones.

TL;DR

Researchers discovered that the "seamlessness" of smart home routines creates a hidden side-channel for attacks. By manipulating shared state variables (like a home's "Away" status), a compromised low-security device (like a smart plug) can trigger routines that disable high-security devices (like a Nest Cam), leading to a novel form of lateral privilege escalation.

What Problem Does This Paper Address?

Most smart home security research focuses on either individual device vulnerabilities or API-based triggers (like IFTTT). This paper targets the Data Store-Based (DSB) paradigm. In DSB platforms, apps and devices don't necessarily talk to each other directly; instead, they read and write to a centralized data store containing state variables (e.g., is_streaming, target_temperature, or a global away flag).

The authors address a critical gap: How does this shared state model affect security? Specifically, they investigate whether the "routines" that make smart homes convenient also make them vulnerable to "confused deputy" style attacks, where a platform is tricked into using its authority to perform unauthorized actions.

Methods

The authors employed a multi-pronged semi-automated analysis:

  1. Permission Mapping: They used the Nest Home Simulator and a physical Hue Bridge to create "permission maps," identifying exactly which data store variables are accessible under specific platform permissions.
  2. Routine Analysis: They extended a smart home setup with 7 real and 2 simulated devices to observe how changes in one variable (e.g., a leak detected by a sensor) triggered actions across the ecosystem.
  3. App Analysis: They performed a security audit on two datasets of Android apps (650 general smart home apps and 39 Nest-specific apps). They checked for:
  4. SSL/TLS Misuse: Using the MalloDroid tool to find broken certificate validation.
  5. Permission Honesty: Comparing the developer-provided descriptions shown to users during OAuth authorization against the actual permissions requested.
  6. Attack Synthesis: They combined these findings to execute a proof-of-concept lateral privilege escalation attack.

Datasets

  • Apps_general: 650 smart home management apps crawled from Google Play.
  • Apps_nest: 39 apps that specifically integrate into the Nest platform.
  • Apps_nest_ext: An expanded set of 111 Android apps built for Nest devices (used for SSL analysis).
  • Physical Devices: Nest Cam, Hue bulb, Belkin Wemo switch, TP-Link Kasa Plug, and others.

Key Insights and Ideas

  • The "Away" Variable as a Global Gadget: In Nest, the away state is a global variable. Even if an app lacks the permission to turn off a camera, it might have permission to set the home to "Home" mode. If the user has a routine that turns off the camera when they return home, the app can indirectly disable the camera by toggling that global flag.
  • Hue’s "Bare Minimum" Failure: Unlike Nest’s fine-grained model, Hue’s access control is woefully inadequate. Once an app is authorized, it can toggle the linkbutton variable to add other malicious apps without the user ever pressing the physical button on the bridge. It can even delete legitimate apps from the whitelist.
  • The Illusion of Product Review: Nest requires a manual product review for apps with >50 users. However, the researchers found that this review is ineffective at checking the content of user prompts. Apps can (and do) use misleading descriptions to trick users into granting high-privilege access.
  • Integrity-Agnostic Execution: The core issue is that DSB platforms treat all writes to the data store as equally "trusted" once authorized, failing to account for the "integrity level" of the source.

Key Results

  • Lateral Privilege Escalation: The authors demonstrated a full attack chain: They compromised a TP-Link Kasa app via an SSL Man-in-the-Middle (MiTM) attack (due to a broken TrustManager). They then used the stolen token to set the Nest home state to "Home," which automatically triggered the Nest Cam to stop recording, allowing for an unmonitored burglary.
  • Widespread SSL Flaws: 20.6% of general smart home apps and 19.8% of Nest-integrated apps contained serious SSL vulnerabilities.
  • Misleading Permissions: 33.3% of analyzed Nest apps had incorrect or misleading permission descriptions. For example, some apps requested "Read/Write" access while their description only claimed they needed to "show" (Read) information.
  • Nest's Technical Correctness: On a positive note, the researchers found that Nest’s internal enforcement of permissions (the mapping of tokens to variables) was technically accurate and fine-grained, unlike many other IoT platforms.

Limitations and Future Work

  • Android Evolution: The SSL MiTM attack was fully functional on Android 4.4.2 but faced hurdles on newer versions (like Android 8.1) due to updated system-level security, though the authors note the underlying app-level vulnerability often remains.
  • Sample Size: The study focused on 39 Nest apps; while significant, it represents only a portion of the "Works with Nest" ecosystem.
  • Defense Proposals: The paper suggests that platforms need Information Flow Control (IFC) to ensure that low-integrity apps cannot influence high-integrity routines. Future work is needed to implement these controls without breaking the "plug-and-play" automation users expect.

Summary updated

Abstract

Home automation platforms provide a new level of convenience by enabling consumers to automate various aspects of physical objects in their homes. While the convenience is beneficial, security flaws in the platforms or integrated third-party products can have serious consequences for the integrity of a user's physical environment. In this paper we perform a systematic security evaluation of two popular smart home platforms, Google's Nest platform and Philips Hue, that implement home automation "routines" (i.e., trigger-action programs involving apps and devices) via manipulation of state variables in a centralized data store. Our semi-automated analysis examines, among other things, platform access control enforcement, the rigor of non-system enforcement procedures, and the potential for misuse of routines. This analysis results in ten key findings with serious security implications. For instance, we demonstrate the potential for the misuse of smart home routines in the Nest platform to perform a lateral privilege escalation, illustrate how Nest's product review system is ineffective at preventing multiple stages of this attack that it examines, and demonstrate how emerging platforms may fail to provide even bare-minimum security by allowing apps to arbitrarily add/remove other apps from the user's smart home. Our findings draw attention to the unique security challenges of platforms that execute routines via centralized data stores, and highlight the importance of enforcing security by design in emerging home automation platforms.
BibTeX
@inproceedings{kafle2019study,
  author = {Kaushal Kafle and Kevin Moran and Sunil Manandhar and Adwait Nadkarni and Denys Poshyvanyk},
  title = {{A Study of Data Store-based Home Automation}},
  booktitle = {CODASPY},
  year = {2019},
  url = {https://openreview.net/forum?id=B-4QhdlrWWq},
}

Loading related content...

Figure