OpenClaw showed us that people really like the idea of a personal assistant but also how incredibly bad things can get without security controls.

@nielsprovos built IronCurtain as a research project to show what a personal assistant might look like if architected with security and reliability in mind.

https://www.provos.org/p/ironcurtain-secure-personal-assistant/

IronCurtain: A Personal AI Assistant Built Secure from the Ground Up

How would you build a personal AI assistant if you took security seriously from the start? IronCurtain sandboxes LLM-generated code, enforces policy in plain English, and keeps credentials out of the agent's reach.

Niels Provos