-
Notifications
You must be signed in to change notification settings - Fork 5.6k
feat(core[exec]): Stronger SandboxVerdict Classifier & Exec Flow #4852
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Open
imajes
wants to merge
3
commits into
openai:main
Choose a base branch
from
imajes:fix/resolve-issues-with-sandbox-denials
base: main
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
+231
−26
Open
Changes from all commits
Commits
Show all changes
3 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The new
stderr_hints_sandbox
function classifies any stderr containing phrases like"permission denied"
or"operation not permitted"
asLikelySandbox
, andprocess_exec_tool_call
turns that verdict into aSandboxErr::Denied
. Many common commands emit those exact strings for reasons unrelated to the sandbox (e.g.,git clone
failing withPermission denied (publickey)
or an SSH server refusing credentials). When such a failure happens under seatbelt/seccomp, the code will now misreport it as a sandbox denial even though the sandbox played no role. This defeats the goal of distinguishing real sandbox denials from normal command failures and will send users down the wrong debugging path. The heuristic should be narrowed or combined with other evidence before upgrading toDenied
.Useful? React with 👍 / 👎.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Well, it's a first step. I agree that there's no such thing as a perfect heuristic; i wonder if we should adopt a weighted approach which assigns each of these "signals" with a weight, and then calculate the probability that it's a sandbox error.
That sounds good, but it becomes super complex. Either way, a
permission denied (publickey)
is likely a sandbox fail as it'll require an escalation to the user to fix.