I gave a (second) talk at #linuxsecuritysummit on a new configuration format, #Landlock Config, designed to define sandboxing security policies in JSON or TOML.
lsseu2025.sched.com/event/25GET

You can easily try it: github.com/landlock-lsm...

Feedback welcome!

AI agents can potentially gain extensive access to user data, and even write or execute arbitrary code.

OpenAI Codex CLI uses #Landlock sandboxing to reduce the risk of buggy or malicious commands: github.com/openai/codex...