Deep Review FAQ

What is deep review?

Deep review is a multi-pass code review workflow that coordinates three specialized subagents — a security reviewer, a quality reviewer, and a test-gap reviewer — and synthesizes their findings into a single consolidated report.

What does the output look like?

The consolidated report contains five sections: Summary (an overall code health score from 0–100 and an executive summary with finding counts by severity), Security, Quality, Test Gaps, and Suggestions (the top 5–10 actionable next steps ordered by impact, each tied to a specific finding).

When should I use deep review instead of a regular code review?

Use deep review when you want all three dimensions — security, code quality, and test coverage gaps — analyzed together in one pass. If you only need one of those dimensions, the focused tools (Security Audit, Bug Predict, or Code Review) are faster and more targeted.

What do I pass to deep_review?

A single path parameter: the path to the directory or file you want reviewed. For example: deep_review(path="src/mypackage").

Which class powers this workflow?

DeepReviewAgentSDKWorkflow in src/attune/workflows/deep_review.py. Call its execute(**kwargs) method to run the workflow and get back a WorkflowResult.

How do I invoke deep review programmatically?

from workflows.deep_review import DeepReviewAgentSDKWorkflow

result = DeepReviewAgentSDKWorkflow().execute(path="src/mypackage")

How do I debug a failing deep review run?

Run the related tests first with pytest -k "deep-review" -v. If the tests pass but your code still fails, add a logger.debug statement at the suspected failure point and re-run with logging enabled. For symptom-based diagnosis, see the troubleshooting page for this feature.

Where is the source code?

src/attune/workflows/deep_review.py

Tags: review, security, quality, tests, comprehensive-review

Unresolved references

Auto-generated by attune-author fact-check. Review and either fix the source code, fix this doc, or add an override.