harness-score

Automate code quality assessment by invoking the evaluation with a specific issue key after the pull request has been merged. The system scores correctness and identifies false positives to update verification reports for reliable deployment decisions.

Run post-merge verification at least 7 days after a feature branch merge to accurately evaluate safety, detect potential regressions, and validate the effectiveness of previous code reviews using internal and external evidence.

Post-merge verdict scoring requires no external dependencies. You only need the specific issue key associated with the merged pull request to automatically evaluate correctness and update verification summaries.

Validate false positives in recently merged code by applying external evidence and internal verdicts to determine validity. This automated retrospective assessment updates quality metrics and verification reports for the team.

Measure code reliability and regressions before deployment by running an automated evaluation that checks if recent merges meet quality standards. It scores the merge's safety and updates verdict summaries to guide deployment.

Post-merge evaluation requires waiting 7 days to allow sufficient time for potential regressions and false positives to surface. This delay ensures the automated quality assessment captures accurate reliability data before updating verification reports.