Your AI agent runs with all your permissions. I built a cage for each tool call, in Rust
This story is from 2026-10-21. It is preserved in the archive; the latest stories are on the live feed.
A subagent got a simple task: read a project's docs and say how to run it. It read docs/readme.txt , found "make run", and then asked for one more file: ~/.ssh/id_rsa . Why it asked doesn't matter. A hidden instruction in something it read, a buggy third-party tool, a model that "thought" the key w…
Read the full story at DEV Community — AI ↗
Timeline · 1 report
- 2026-10-21 20:33 · DEV Community — AI
Your AI agent runs with all your permissions. I built a cage for each tool call, in Rust