Summary
A flaw in Home Assistant's Konnected integration leaves a read-only HTTP endpoint accessible without authentication, exposing sensor data and configuration details to anyone who can reach the service.
Because GET requests skip access token verification while write requests are protected, attackers on the same network or with routed access can query sensitive state information without needing credentials.
Affected context
This vulnerability occurs in Home Assistant deployments that enable the Konnected integration and expose the HTTP interface over a network reachable by untrusted parties.
The KonnectedView endpoint is explicitly marked as not requiring authentication, and while write operations enforce an access token, read (GET) requests bypass any authentication checks.
Environments with remote access, port forwarding, reverse proxies, or shared local networks are particularly at risk, as they broaden who can reach the unauthenticated endpoint.
Exploit preconditions
- Attack surface: HTTP KonnectedView endpoint in Home Assistant
- Required access: Network-level ability to send GET requests to the Home Assistant instance (e.g., same LAN, VPN, or exposed port)
- Trigger point: Issuing a GET request to the KonnectedView path without providing valid authorization headers
- Why exploitation is possible: The GET handler (get()) omits any authentication or token validation, despite the endpoint being network-accessible and assumed to be protected by configuration-based access tokens
Why this happens
The underlying issue is a mismatch between the intended authentication model and the actual implementation of HTTP handlers.
The KonnectedView endpoint is globally marked as not requiring authentication, with a design assumption that all operations would be protected "via the access token from configuration."
In practice, only write operations (POST and PUT) call update_sensor() with proper Authorization header checks, while read operations are handled by a separate get() method that performs no authentication.
This divergence maps to CWE-200: Exposure of Sensitive Information to an Unauthorized Actor and CWE-306: Missing Authentication for Critical Function, where a critical data retrieval path is left unguarded.
Potential impact
- Unauthorized reading of sensor values and device state information
- Exposure of environmental and occupancy data that could aid physical intrusion or stalking
- Leakage of integration-specific configuration details that may assist further attacks
- Increased attack surface for reconnaissance against the Home Assistant instance
General mitigation guidance
Mitigation generally focuses on enforcing consistent authentication and minimizing unauthenticated data exposure.
Common patterns include:
- Ensuring all HTTP handlers, including read-only endpoints, perform authentication and authorization checks aligned with the integration's access token model
- Centralizing auth enforcement in shared middleware or view base classes to avoid divergent behavior between GET and write methods
- Reducing the amount of sensitive information returned by any endpoint, especially those reachable without strong authentication
- Constraining network exposure of home automation services through firewall rules, segmented networks, and TLS-terminating gateways following least-privilege principles
How teams detect issues like this
Detection can rely on a combination of code-level and deployment-level checks:
- Static analysis rules or linting that flag view or route registrations marked as unauthenticated while still handling sensitive data
- Code review practices that_focus on verifying authentication and authorization logic for all HTTP verbs (GET, POST, PUT, DELETE) exposed by an integration
- Automated tests that exercise endpoints without credentials to confirm appropriate rejection of unauthenticated requests
- Configuration audits to identify home automation services exposed to untrusted networks and to validate adherence to internal security policies for external access

