Claude-Code Automode

#1

Post by marginalx » Tue, Mar 24, 2026, 6:03 PM UTC

Re: Claude-Code Automode

#2

Post by daisydevel » Tue, Mar 24, 2026, 6:15 PM UTC

Nice to have a safer alternative to --dangerously-skip-permissions. But unless it's a critical project, I'll probably just keep using --dangerously-skip-permissions until I get comfortable with the new approach...

Re: Claude-Code Automode

#3

Post by dakolli » Tue, Mar 24, 2026, 11:31 PM UTC

Its funny how this has one comment and seven points on HN but 24k likes and 1.4k replies on Twitter.

Definitely a great indication of how manic and hyped Twitter (X) is (or how much easier it is to buy engagement)..

https://x.com/claudeai/status/2036503582166393240

Re: Claude-Code Automode

#4

Post by 1123581321 » Wed, Mar 25, 2026, 2:06 AM UTC

A classifier is probably nice for those who currently allow --dangerously-skip-permissions, but it's not for those who have been trying to only allow the right commands to always run. It only lowers the odds of something bad happening. Maintaining a massive allowlist that parses nested bash commands is safer. (I do this. It fits in a 2MB binary that runs on a hook, and it includes what I've put in Claude's allowlist after parsing and tokenizing nested bash.)

Re: Claude-Code Automode

#5

Post by splitbrain » Wed, Mar 25, 2026, 2:36 AM UTC

1123581321 wrote: post #4

A classifier is probably nice for those who currently allow --dangerously-skip-permissions, but it's not for those who have been trying to only allow the right commands to always run. It only lowers the odds of something bad happening. Maintaining a massive allowlist that parses nested bash commands is safer. (I do this. It fits in a 2MB binary that runs on a hook, and it includes what I've put in Claude's allowlist…

Is that available somewhere? Maybe as a blog post on how you set this up?

Re: Claude-Code Automode

#6

Post by 1123581321 » Wed, Mar 25, 2026, 11:24 AM UTC

splitbrain wrote: post #5

1123581321 wrote: post #4

A classifier is probably nice for those who currently allow --dangerously-skip-permissions, but it's not for those who have been trying to only allow the right commands to always run. It only lowers the odds of something bad happening. Maintaining a massive allowlist that parses nested bash commands is safer. (I do this. It fits in a 2MB binary that runs on a hook, and it includes what I've put in Claude's allowlist…

Is that available somewhere? Maybe as a blog post on how you set this up?

I have it on Github. https://github.com/michaeldhopkins/safe-chains

All of the commands and flags are listed here. https://github.com/michaeldhopkins/safe-chains/blob/main/COM...