CVE-2026-50083
Description
A hardcoded OAuth client secret in the Aqara IAM/SSO gateway lets unauthenticated attackers obtain a 57-year valid token with full scope.
AI Insight
LLM-synthesized narrative grounded in this CVE's description and references.
A hardcoded OAuth client secret in the Aqara IAM/SSO gateway lets unauthenticated attackers obtain a 57-year valid token with full scope.
Vulnerability
The Aqara IAM/SSO Gateway (gw-builder.aqara.com) contained two hardcoded OAuth client credentials (client_id test1 with secret 123456, and client_id test123 with secret 172799) that were accepted by the /iam/oauthToken/openapi/client/token endpoint. These credentials, an instance of CWE-798, allowed issuance of tokens with scope=all and an expiry of up to 1,799,999,999 seconds (approximately 57 years) [1][2]. The affected versions include all releases prior to the vendor's fix on 2026-04-20.
Exploitation
An attacker with network access to gw-builder.aqara.com can obtain an access token by sending a simple POST request containing the hardcoded client_id and client_secret with grant_type=client_credentials. No authentication or user interaction is required [2]. The token is immediately valid. The IAM gateway does not rate-limit token issuance [2].
Impact
Successful exploitation gives the attacker a bearer token with scope=all, enabling any privileged API call against the IAM gateway. When chained with CVE-2026-50084 (application signing bypass) and CVE-2026-50082 (auth code delivery), this allows unauthenticated, remote full compromise of Aqara devices including smart locks, cameras, and hubs [1][2].
Mitigation
The vendor remediated the issue by 2026-04-20, removing the hardcoded credentials from the gateway [2]. Users should ensure their Aqara account and devices are updated. No workaround is available for unpatched instances. The vulnerability is not listed on CISA's KEV as of the publication date.
AI Insight generated on Jun 12, 2026. Synthesized from this CVE's description and the cited reference URLs; citations are validated against the source bundle.
Affected products
1Patches
0No patches discovered yet.
Vulnerability mechanics
No source-code context for this CVE — mechanics is only generated when we can read the actual fix diff. Without that, the four sections (root cause, attack vector, affected code, fix) would be speculation rather than analysis.
References
1News mentions
0No linked articles in our index yet.