Trust & Evaluation
Methodology & Benchmarks
This page explains how Zush evaluates AI file renaming quality for macOS workflows. We publish the rubric, validation flow, and update cadence so recommendations stay reproducible, auditable, and practical for real batch operations.
Editorial Accountability
Zush Editorial Team
Zush Product Team
Criteria are reviewed monthly, with material changes logged in the changelog.
Scoring Rubric
Each workflow run is graded against a weighted model that balances naming accuracy, operational safety, and day-to-day usability inside a macOS workflow.
| Dimension | Weight | What we measure |
|---|---|---|
| Semantic filename accuracy | 35% | Does the new name describe file meaning, not only generic patterns? |
| Consistency in batch runs | 20% | Stable naming style across mixed files in one task. |
| Automation depth | 15% | Batch workflow, folder monitoring, and repeatability. |
| Safety controls | 15% | History, rollback, preview-before-apply behavior. |
| Operational fit | 15% | macOS workflow quality, speed, and setup friction. |
Benchmark Protocol
- 01
Use a mixed-file benchmark set with screenshots, photos, PDFs, and office documents.
- 02
Run controlled rename jobs with identical prompts and naming constraints.
- 03
Score outputs against the rubric and record false positives or ambiguous labels.
- 04
Validate revert and recovery behavior before marking a setup as production-safe.
- 05
Repeat checks after release updates and track material changes in the changelog.