[jeromevdl]

This is a pre-PR for automated reviews. It gives human reviewers—and a future automated reviewer following AGENTS.md—deterministic signals about risky changes. Automated review itself will follow separately.

• Compare function complexity and backend branch coverage against the merge base, with direct production-importer counts to highlight widely used files.
• Publish one updated, advisory PR comment. Omit unchanged complexity debt; label missing or partial measurements explicitly.
• Skip backend coverage suites for unrelated changes, reuse the correct artifacts on selective reruns, and preserve the report if dependency analysis fails. Fork/Dependabot comments use validated artifacts and default-branch publishing code.

Validation: all 6,058 backend tests and 42 focused quality-report tests passed, along with lint, formatting, secret scanning, and workflow validation. A local 60-commit benchmark improved from 45.2 s to 34.7 s with identical reports. Full script suite: 136 passed; two existing local failures came from scratch-package metadata and the unrelated, untracked Terraform test (excluded from this PR). Hosted workflow execution remains to be verified by this PR.

[github-actions]

Complexity and test coverage

Analyzed 0 source files changed by this PR. Advisory threshold: 15.

0 new/unmatched or worsening functions above the threshold.

🟢 Improving · 🟠 Needs attention · 🔴 High risk · 🟣 Highest risk

unknown = unavailable · n/a = no branches · * = incomplete tests

CI run and full report artifacts

[JWThewes]

That's great!