AI Review tool
The AI Review tool asks a configured LLM provider to review each translation and return a concise assessment: a score from 1 to 10, a brief judgement of accuracy and fluency, and a suggested improvement where warranted. The review text is recorded on the block's properties along with the provider name. Blocks without a translation for the target locale are skipped.
Both a target locale and provider credentials are required, since the tool makes an external API call per block. Unlike the AI QA check, which returns structured issues, this tool produces a free-form review intended for human reviewers deciding whether a translation needs revision.
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
apiKey | string | API key for the AI provider | |
model | string | AI model name | |
provider | string | anthropic | AI provider |
Configure these parameters interactively and copy the flow-step YAML on the Tool Reference.
Examples
Review with the default provider
Score and assess each translation using Anthropic.
provider: anthropic
Review with a specific model
Use a named OpenAI model for the review.
provider: openai model: gpt-4o
Processing notes
Operates on blocks that have a translation for the target locale.
Blocks are processed in parallel by default for throughput.
The review text and provider name are stored on the block's properties.
Limitations
Requires AI provider credentials and a target locale; makes an external API call per block.
The review is free-form model output, not structured QA issues.
Blocks without a translation for the target locale are skipped.
← Back to the Tool Reference