feat(autofix): Thread insert_index through explorer API for step retry#112287
Merged
isaacwang-sentry merged 5 commits intomasterfrom Apr 7, 2026
Merged
Conversation
Add a new Flagpole feature flag to gate "Retry from step" buttons on Autofix v3 cards. This is primarily useful for local Seer testing, allowing developers to re-run individual autofix steps without restarting the entire flow. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
github-actions
bot
added
the
Scope: Backend
Pass insert_index from the API serializer through trigger_autofix_explorer to client.continue_run. When provided, Seer truncates blocks after that index, enabling retry-from-step to properly reset downstream steps. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Contributor
Backend Test FailuresFailures on
|
The retry buttons are no longer gated behind a feature flag, so this registration is unnecessary. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
isaacwang-sentry
changed the title
Contributor
Backend Test FailuresFailures on
|
The test_stopping_point test checks exact kwargs passed to trigger_autofix_explorer. Add the new insert_index=None parameter to match the updated function signature. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
aliu39
approved these changes
Apr 7, 2026
Verify that insert_index from the POST request is threaded through to trigger_autofix_explorer for retry-from-step functionality. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit f299f7a. Configure here.
| prompt_metadata=prompt_metadata, | ||
| artifact_key=artifact_key, | ||
| artifact_schema=artifact_schema, | ||
| insert_index=insert_index, |
Contributor
There was a problem hiding this comment.
insert_index silently ignored when run_id is absent
Low Severity
When a caller provides insert_index without run_id, the parameter is silently dropped because trigger_autofix_explorer only passes insert_index to client.continue_run() in the else branch (existing run), not to client.start_run(). The endpoint has no validation to reject this combination — unlike open_pr and coding_agent_handoff, which explicitly check that run_id is present. A retry-from-step request missing run_id would quietly start a brand-new run instead of truncating and retrying.
Additional Locations (1)
Reviewed by Cursor Bugbot for commit f299f7a. Configure here.
isaacwang-sentry
deleted the
isaacwang-sentry/feat/autofix-retry-from-step-flag
branch
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Thread
insert_indexthrough the explorer autofix API so the frontend can requesta retry-from-step that truncates all downstream blocks. When provided, Seer does
blocks = blocks[:insert_index]before re-running the step, ensuring downstreamsections (e.g. Code Changes after retrying Plan) are properly removed.
Changes:
insert_indexfield toExplorerAutofixRequestSerializertrigger_autofix_explorer()→client.continue_run()A follow-up frontend PR (#112290) adds retry buttons that use this parameter.