Summary
The Home Assistant iOS Companion app may treat tag links delivered through iOS routing mechanisms as legitimate NFC or QR tag scans without verifying their origin or asking the user for confirmation. An untrusted app on the same device could therefore trigger an authorized automation silently and without the user's interaction.
Affected context
The issue occurs in Home Assistant iOS deployments that process NFC or QR tag links through operating-system routing mechanisms such as iOS universal links. It becomes risky when the application accepts a forwarded tag as equivalent to a physical scan without authenticating the calling application, validating the delivery context, or requiring user confirmation.
Exploit preconditions
- Attack surface: The Home Assistant iOS Companion app's tag-link handling and iOS routing integration
- Required access: An untrusted application installed on the same iOS device, with the ability to forward or invoke a tag link
- Trigger point: Processing of the forwarded NFC or QR tag link
- Why exploitation is possible: The app accepts the routed tag as an authorized physical scan without caller validation or interactive confirmation
Why this happens
The underlying bug is an authorization failure in which the application trusts an externally delivered event without establishing that it originated from an authorized physical interaction or trusted caller. This maps to CWE-862: Missing Authorization, because a security-sensitive automation action is executed without checking whether the source is permitted to invoke it. The insecure pattern is treating an inter-application message or deep link as equivalent to a trusted user action.
Potential impact
- An untrusted local application may trigger automations without user awareness
- Security-sensitive devices or services may be controlled unexpectedly
- Home occupants, property, or connected equipment could be exposed to safety or privacy risks
- Attackers may cause repeated or timed automation actions through local application activity
General mitigation guidance
Mitigation generally focuses on authenticating the source and context of tag-triggered requests before executing automations.
Common defensive patterns include:
- Distinguishing verified physical scans from externally routed links
- Validating the calling application and the integrity of the delivery context
- Requiring explicit user confirmation for sensitive or destructive automations
- Applying per-automation authorization and least-privilege controls
- Limiting which tags can invoke high-impact actions
- Treating deep links and inter-application messages as untrusted input by default
How teams detect issues like this
Organizations can look for authorization gaps in mobile deep-link and inter-application message handlers through:
- Static analysis rules covering externally invoked routes that directly trigger privileged actions
- Code review checks for missing caller validation, origin checks, or user-confirmation gates
- Test cases that invoke tag handlers through universal links and other OS routing paths
- Runtime telemetry correlating automation execution with physical scan events and application identity
- Mobile application policies requiring authentication and explicit consent before high-impact actions
Automated review tools may help identify routes where untrusted link data reaches automation execution without an authorization decision.

