CommandCodeAI / CommandCodeAI/desktop
[Feedback]: Have proper sandboxed execution logic
@aliarain is already working on this.
Since Sep 7, 2026.
- Dominant language
- Shell
- Stars
- 80
- Forks
- 2
- PR merge metrics
- No merged PRs in 30d
Description
What problem are you trying to solve?
The current permission settings are not sufficient for my use case. I general, the available approval modes are hard for me to understand.
What would improve it?
What I want is sandboxed execution (I am on Windows). A bit like in Codex (potentially the only app that got this right; GitHub Copilot in VS Code is not doing this, neither is Claude Code. They have a sandbox. Reads are typically allowed. Writings outside the project folder, internet access, ... requires permission. And because they have a sandbox, they can also apply this to arbitrary command executions. They can ensure that nothing can be circumvented, or hidden inside some long script, because they always limit the real effect with their sandbox.
What do you do today?
No response
Product area
None
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.