Reproduce the AI Reviewer's Claim Before You Fix It

Описание и применение

Внедрение порога воспроизводимости проверяет выводы бесплатных AI‑ревьюеров, снижая ложные тревоги и экономя время разработчиков.

AI reviewers on free tiers can produce inaccurate security findings that mislead developers, causing wasted effort or incorrect fixes.

Implementing a reproducibility gate reduces false positives from free AI reviewers, preventing incorrect code changes and saving engineering time.

Pin the reviewer's world by setting a specific commit and Docker image for the review.
Run the AI reviewer against the pinned state and capture structured findings in JSON.
Require each finding to include an executable evidence block that can be run locally.
Execute the evidence scripts; if all pass, mark the claim as verified; otherwise flag as fail or no-info.
Let a human decide whether to apply the fix based on the gate outcome.

Free quota is a great reason to try a platform. It is not a reason to trust its output.

The problem is rarely "the model is 100% wrong." The problem is "the model is 30% wrong, and the wrong 30% looks identical to the right 70%."