patent-translator

To translate English patent documents to Korean with consistent terminology, use a four-agent orchestration workflow that analyzes, translates, reviews, and learns from sections like TAC, Background, and Claims while maintaining a dynamic term base.

Yes, you can export Korean patent translations directly to a Word docx file. The workflow uses the python-docx package to automatically compile and format output with Batang 12pt font and justified text.

A dynamic terminology base in patent translation is a real-time term accumulation process that captures and enforces consistent vocabulary across TAC, Background, Drawings, and Detailed Description sections during the translation workflow.

You ensure legal-grade quality control in Korean patent translation by enforcing a 95-point quality threshold that combines deterministic mechanical checks with semantic review to validate accuracy and compliance.

Yes, you need Python 3.8 or higher with the python-docx and lxml packages, along with the Claude Task/Subagent framework, to orchestrate the analysis, translation, review, and learning loop for the patent translation workflow.