You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Preserve workbench availability when sandbox API responses have no recorded provenance; record that condition without discarding existing secret history.
Reuse file and table delivery provenance and remove the catalog allowlist and response-copying check. Keep explicit unknown-input handling and existing redaction rules.
Let table mutations finish before recording response provenance; withhold an undeliverable response with its settled HTTP outcome instead of interrupting completion.
Type of Change
Bug fix
Testing
32 live Redis/local-process checks and 61 focused tests passed. Confirmed the new regressions fail before the fix and the table redaction check fails without its admission guard. Application type-check, lint, all repository audits, standard generators, staging-aware block registry, and docs manifest checks passed. Detailed evidence stays local.
Checklist
Code follows project style guidelines
Self-reviewed my changes
Tests added/updated and passing (new tests pass the test-audit authoring gate)
[Medium risk] Adjusts sandbox resource transport and test coverage.
The PR appears safe to merge; no new actionable failure was established, and the prior mutation-completion finding is fixed.
Summary
The PR keeps the workbench available after API responses without recorded provenance and defers table-row provenance admission until the API operation completes.
File and table delivery observers continue to supply explicit evidence.
If table-row provenance cannot be stored, the transport withholds the completed response body and reports its settled HTTP outcome.
Diagram
sequenceDiagram
participant S as Sandbox
participant T as Resource transport
participant A as API handler
participant P as Provenance storage
S->>T: API request
T->>A: Dispatch request
A-->>T: Completed response and delivery evidence
T->>P: Record row provenance, if observed
alt Recording succeeds
T-->>S: Original response
else Recording fails
T-->>S: 502 with body withheld and completed outcome
end
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.
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
Summary
Type of Change
Testing
32 live Redis/local-process checks and 61 focused tests passed. Confirmed the new regressions fail before the fix and the table redaction check fails without its admission guard. Application type-check, lint, all repository audits, standard generators, staging-aware block registry, and docs manifest checks passed. Detailed evidence stays local.
Checklist