Claude Code's "Don't Ask Again" exposes passwords in 1,122 public GitHub files

Re:Claude Code's "Don't Ask Again" exposes passwords in 1,12... by kstenerud on Saturday November 07, 2026 @10:53PM (#50410284)

This is EXACTLY why I never use an agent outside of https://github.com/kstenerud/yoloai

I appreciate that you're trying to make Claude Code safer, but you've been doing it the wrong way.

The agent should NOT have ANY access to:

- Your ENV

- Your homedir

- The network at large

- ANY other directory on your system

- No, not even your workdir.

The only safe way to use an agent is COMPLETELY isolated by a container or VM, using git semantics to extract its output.

[ Parent | on HN ]

Re:Claude Code's "Don't Ask Again" exposes passwords in 1,12... by post-it on Saturday November 07, 2026 @11:34PM (#50410442)

That's the only safe way to use any software that you haven't thoroughly reviewed. But that's not how I use any software.

[ Parent | on HN ]

Re:Claude Code's "Don't Ask Again" exposes passwords in 1,12... by kstenerud on Saturday November 07, 2026 @11:47PM (#50410487)

You could in theory fully review Claude Code.

But even that wouldn't be enough because LLMs are nondeterministic, and these kinds of issues are the inevitable results.

Only a mechanical, deterministic failsafe can protect you from it.

[ Parent | on HN ]

Re:Claude Code's "Don't Ask Again" exposes passwords in 1,12... by what on Sunday November 08, 2026 @12:51AM (#50410762)

> You could in theory fully review Claude Code.

Can you point me to the CC source code?

[ Parent | on HN ]

Re:Claude Code's "Don't Ask Again" exposes passwords in 1,12... by kstenerud on Sunday November 08, 2026 @01:03AM (#50410819)

The CC source code leak was contained many months ago. But Claude is perfectly happy to decompile its own code if you ask the right way. It'll even patch the binary for you.

Regardless, It's kinda silly to keep the harness code hidden. There aren't any valuable secrets in there.

But there are rules & procedures to follow, so...

[ Parent | on HN ]

Re:Claude Code's "Don't Ask Again" exposes passwords in 1,12... by what on Sunday November 08, 2026 @01:11AM (#50410851)

Well then I have to use Claude before being able to review the source? Bit of a chicken and egg situation.

I also have to trust its “decompilation” is accurate.

[ Parent | on HN ]

Re:Claude Code's "Don't Ask Again" exposes passwords in 1,12... by kstenerud on Sunday November 08, 2026 @01:19AM (#50410888)

You're absolutely right! There's no way around this problem — and no amount of creative thought could ever solve it.

All hail the agent!

[ Parent | on HN ]

Re:Claude Code's "Don't Ask Again" exposes passwords in 1,12... by codedokode on Sunday November 08, 2026 @12:29AM (#50410667)

I have no time to review software. I want Linux to be able to run any program without any risk. Sadly it cannot so I have to resort to ugly solutions like VM. Still better than nothing.

[ Parent | on HN ]

Re:Claude Code's "Don't Ask Again" exposes passwords in 1,12... by saidnooneever on Sunday November 08, 2026 @02:55AM (#50411259)

its pretty trivial in any harness just to not have it post your shit to the public internet automatically. its a dumb idea to begin with

[ Parent | on HN ]

Re:Claude Code's "Don't Ask Again" exposes passwords in 1,12... by grey-area on Sunday November 08, 2026 @06:56AM (#50412496)

Text in a model context is not a reliable constraint.

[ Parent | on HN ]