|
Documentation
Book a DemoPlatform
PlatformMCPCLIAPIWorkflows
GuidesChangelog
All releases

W29 – AI review false positives down 59.7%

AI reviews got more accurate this week. On our internal benchmark, false positives dropped by 59.7%. A false positive is when a review marks a correct translation as failing. Fewer false positives means fewer alerts to check by hand, and review scores you can trust more.

Under the hood, we made AI Reviews more accurate. The biggest change: when a review doesn't apply to a given string, it no longer counts against the score. Before, a "not applicable" result could be treated as a failure and pull the number down, so translations looked worse than they actually were. Scores now reflect only the reviews that genuinely apply.

The reviewer setup form now shows example instructions as a placeholder for each review type.

The GitHub App localizes Xcode String Catalogs#

The GitHub App now handles .xcstrings files. Commit a String Catalog to a connected repository and every push localizes it through your engine's pipeline, the same way the CLI and Async Localization API do.

Also shipped#

  • Source file path in request logs. Request logs from lingo push runs now record which file each request came from, so a log entry traces back to the file that produced it.

Fixed#

  • Key names now guide translation in JSON and JSONC files. Translations now use your real key names as context, so a key like checkout.error.cardDeclined informs the result instead of the value being translated by its position in the file alone. Applies automatically across the CLI, the GitHub App, and the Async Localization API.
Veronica PrilutskayaVeronica Prilutskaya, CPO & Co-Founder·Published 3 days ago·1 min read