Key points
- CVE-2026-110616 is rated Critical with a CVSS score of 9.2.
- The vulnerability exists in nullclaw NullHub versions through 2026.5.29.
- Attackers can control agents, rewrite credentials, and manage cron jobs.
Unauthenticated attackers can now take full control of nullclaw NullHub agent instances. This capability arises because the software fails to enforce authentication on its management API. According to the National Vulnerability Database, the flaw allows remote users to access sensitive endpoints without providing valid credentials.
In plain English
The core issue is a missing authentication check. The system’s `auth_token` field defaults to null and is never assigned a value. Consequently, attackers sending requests without Authorization or Origin headers to non-loopback listeners can access the API. This design choice effectively leaves the back door open for anyone who can reach the server.
The background
This vulnerability affects nullclaw NullHub up to version 2026.5.29. It is tracked as CVE-2026-110616 and classified under CWE-306, which covers missing authentication for critical functions. The NVD record describes it as a critical severity issue, reflecting the ease of exploitation and the high impact on confidentiality, integrity, and availability.
What changes now
Organisations using affected versions face immediate risk if their instances are accessible from non-loopback networks. Attackers can rewrite provider credentials, manage cron jobs, and control MCP servers. The vulnerability allows full agent control. No patch has been confirmed yet.
The attacker's path
Teams usually encounter this class of flaw when API endpoints lack strict identity verification. The weakness depends on default configurations that fail to reject unauthenticated requests. This technique serves the initial access stage of an intrusion, allowing attackers to establish a foothold without brute-forcing credentials. Check whether your APIs enforce authentication on all management routes. Verify that default tokens are not null or empty. Ensure that internal services are not exposed to untrusted networks without additional access controls. This approach helps mitigate the risk of unauthorised configuration changes.
What to do and how to stay safe: nullclaw
- Isolate affected nullclaw NullHub instances from untrusted networks until a fix is available.
- Monitor logs for requests to management APIs lacking valid authentication headers.
- Review agent configurations for unexpected changes or new cron jobs.
- Restrict access to API endpoints using network-level firewalls or allowlists.
Step-by-step guide: Emergency Patching: Correcting Five Common Operational Errors
General security guidance from the RedTeam Log newsroom. It is not confirmed advice from the organisations named in this story.
Frequently asked questions
What is the CVSS score for this vulnerability?
The CVSS score is 9.2, which is classified as Critical.
Which versions of nullclaw NullHub are affected?
All versions of nullclaw NullHub up to and including 2026.5.29 are affected.
Can attackers access the API without credentials?
Yes, the vulnerability allows unauthenticated access to the management API.
