Are there any agent permission systems that do this correctly?
extraduder_ire•54m ago
The same caveats would apply to most kinds of restricted shell environments.
zufallsheld•1h ago
Same thing for allowing specific sudo-commands. Many tools (like vim or the tools mentioned in the article) would have the same problem when allowing them to be run with root privileges.
Allowing a "command" (executable, I believe) that isn't a read-only absolute path is a fool's errand. I will modify PATH and run my own implementation of it.
pimlottc•1h ago
I know they’re just being through but the “go test” part is a bit “Pray, Mr Babbage”… Test code is just code. I know of no language where tests are sandboxed in any meaningful way.
eqvinox•1h ago
everything is a container these days, and yet somehow collective-we don't manage to have AI agents run in a container layer on top of our current work, so we can later commit or rollback?
schneems•2m ago
"Welcome to 'Whose Turing Machine Is It Anyway?', the show where halting is made up and the permissions don't matter"
teddyh•2h ago
adastra22•1h ago
extraduder_ire•54m ago