Skip to content

Forum

AI Assistant
Forums
OpenClaw
Sandboxing and Exec...
Sandbox Escapes and...
 
Notifications
Clear all

Sandbox Escapes and Breakout Research

Known and theorized sandbox escape paths in OpenClaw — research, CVE discussion, and responsible disclosure coordination. Claims should come with reproducible steps or clear reasoning.
RSS
Page 2 / 2

Is the agent's memory system a viable escape route?

Last post by Pia Voss, 5 days ago

Posts: 9
Views: 2


Kubernetes Pod Security Context vs custom container - which is safer?

Last post by Oscar Lindqvist, 6 days ago

Posts: 12
Views: 2


My results after running OpenClaw under Landlock - partial success

Last post by Omar F., 6 days ago

Posts: 3
Views: 2


Thoughts on the proposed 'capability-based' security model in the RFC?

Last post by Anna L., 6 days ago

Posts: 8
Views: 7


Am I the only one who thinks the sandbox docs overstate its capabilities?

Last post by Pete Audits, 7 days ago

Posts: 10
Views: 2


Help: Agent keeps spawning child processes I didn't authorize

Last post by Tom Mod, 1 week ago

Posts: 1
Views: 2


Unpopular opinion: If you need a sandbox, your agent design is already flawed.

Last post by Viktor Petrov, 1 week ago

Posts: 1
Views: 6


Page 2 / 2
Share:
Share
Tweet
Share