reocr

Name: reocr
Availability: InStock
Author: strigov

Reprocess poorly OCRed legal documents with Haiku Vision and update mirrors and indices.

5|Updated Mar 30, 2026

Search Tags:#ocr #vision #document-processing #haiku #legal-analysis #evidence-management #mirror-update

Ready to run

npx skills add https://github.com/strigov/vassal-litigator --skill reocr

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: reocr
Source: https://github.com/strigov/vassal-litigator/tree/main/skills/reocr
Command: npx skills add https://github.com/strigov/vassal-litigator --skill reocr

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Повторное OCR плохо распознанных юридических документов через Haiku vision, чтобы повысить точность текста и корректно обновлять зеркала дела и индекс материалов.

Core Features & Use Cases

Повторная обработка документов с низким качеством OCR через Haiku vision
Обновление зеркал и индекса .vassal/index.yaml после обработки
Поддержка документов, найденных в индексе дела, с автоматическим повторным OCR

Quick Start

Запустите повторное OCR для документов с низким качеством распознавания через Haiku vision.

Frequently Asked Questions about reocr

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I re-OCR legal documents that have poor text accuracy?▼

Re-OCR for legal documents is performed by rendering pages and triggering a Haiku vision subagent to reprocess poorly recognized text, then automatically updating case mirrors and the index.

What is the best way to improve OCR quality for evidence management files?▼

Improving OCR quality for evidence files involves using Haiku vision to reprocess document pages, correcting text errors, and updating the `.vassal/index.yaml` to reflect the improved materials.

Do I need a specific renderer script to re-OCR documents with Haiku vision?▼

Yes, re-OCR with Haiku vision requires a working renderer script to render document pages, Haiku vision access, and proper path validation to keep documents within the case root.

Can I reprocess documents that are already tracked in a case index?▼

Yes, documents tracked in the `.vassal/index.yaml` case index are automatically targeted for re-OCR, rendering pages and updating mirrors and indices after processing.

How does mirror update work after re-OCRing legal documents?▼

Mirror updates occur automatically after Haiku vision reprocesses document pages, synchronizing the improved text accuracy with case mirrors and the `.vassal/index.yaml` material index.

Why does path validation matter when re-OCRing case documents?▼

Path validation ensures documents stay within the case root during re-OCR, preventing errors when rendering pages and updating mirrors and indices for legal evidence management.