Inspect AI writing scores and feedback before changing your model or prompt. This free preview contains a six-page PDF, three synthetic writing cases with provisional references, a blind JSONL input file and a short start guide.

Use it to explore task fulfilment, language evidence and feedback accuracy, then inspect the full repeatable QA workflow in AI Score Benchmark Pro.

Synthetic Development Edition v1.0.0. Responses and ratings are AI-authored provisional design material. Independent human rating has not been completed. This sample supports exploratory internal QA; it is not an operationally validated human benchmark or an official CEFR assessment. A total out of 25 does not convert to CEFR.

Full toolkit: $199 USD one-time. 92 synthetic records, two editable XLSX workbooks, structured exports, an offline Python evaluator, a local report viewer and buyer guides. Team Evaluation Licence: one organisation, up to five named internal users. Bring your own model outputs.

View the full toolkit: https://contra.com/products/0caiR8A2-ai-score-benchmark-pro-writing-assessment-qa-kit

Use the preview to inspect the format, then send your rubric, intended learners and main scoring concern through Contra. The AI Writing Assessment and Feedback Quality Audit is a separately booked review of client materials.