AutoAgent TCP server unauthenticated remote code execution as root
AutoAgent's TCP server binds to all network interfaces by default and executes attacker-supplied commands without authentication, allowing remote attackers to run arbitrary code as root.
- Vendor
- HKUDS
- Product
- AutoAgent
- CVSS
- 9.8
- EPSS (exploit probability)
- 1.0%
- Status
- patched
- Published
Is CVE-2026-86124 being exploited?
Not on any exploited-vulnerability catalog we track. CVE-2026-86124 was not on the CISA, ENISA or CIRCL exploited catalogs as of our 2026-10-23 snapshot. That can change quickly; check the write-up below for the latest reporting.
FIRST's EPSS model puts the probability of exploitation activity in the next 30 days at 1.0% (scored October 14, 2026).
- CISA KEV[ — ]
- ENISA EUVD (exploited)[ — ]
- CIRCL KEV[ — ]
Catalog snapshot 2026-10-23. Full cross-catalog view: /kev-comparison
CVE-2026-86124 timeline
Every event links to its source. Catalog dates are each catalog's own; day offsets count from NVD publication. Data: JSON
AutoAgent, an AI agent framework maintained by HKUDS, contains an unauthenticated remote code execution vulnerability in its TCP server component. The server binds to all network interfaces by default and accepts commands from any connecting client without authentication. Commands are executed as root.
The flaw affects AutoAgent installations exposed to untrusted networks. Default deployments are vulnerable without network-level controls. Check the GitHub advisory for the fixed version. Apply the patch or restrict TCP access to trusted sources via firewall rules if the patched version is not yet available.
