Forum

Forums
OpenClaw
Architecture and Th...
Trust Boundaries an...
 
Notifications
Clear all

Trust Boundaries and Component Isolation

How OpenClaw separates the orchestrator, tool executor, and model backend — and what happens when those boundaries break. For developers and appsec engineers mapping lateral movement risk.
RSS
Page 1 / 5

Breaking: OpenClaw team publishes a threat model — finally!

Last post by Emma Clarke, 4 weeks ago

Posts: 6
Views: 2


My results after stress-testing the model backend with malicious tool outputs

Last post by Priya Mendis, 4 weeks ago

Posts: 1
Views: 3


Am I the only one who finds the default OpenClaw config too permissive?

Last post by Nina Osei, 1 month ago

Posts: 3
Views: 2


Has anyone tried wrapping the model backend in a microVM?

Last post by Mia Chen, 1 month ago

Posts: 3
Views: 2


Switched from AutoGen to OpenClaw because of the explicit trust boundary documentation

Last post by Julia Sterling, 1 month ago

Posts: 2
Views: 1


Guide: Using AppArmor to lock down the tool executor’s system calls

Last post by Grace W., 1 month ago

Posts: 1
Views: 1


Complete newbie here — what tools do I need to start securing my OpenClaw install?

Last post by Ken Cloud, 1 month ago

Posts: 1
Views: 2


Anyone else having issues with the orchestrator not respecting network egress rules?

Last post by Nick R., 1 month ago

Posts: 2
Views: 2


Did you see the vulnerability disclosure about orchestrator-to-executor RCE?

Last post by Morgan Lee, 1 month ago

Posts: 2
Views: 2


Did you see the talk about lateral movement risks in agent runtimes?

Last post by Ivan Sokolov, 1 month ago

Posts: 1
Views: 2


Page 1 / 5
Share:
Share
Tweet
Share