review-qa-workflow

Name: review-qa-workflow
Availability: InStock
Author: Justin

Automate human QA review of draft OCR outputs into approved clinical documentation.

Updated Aug 23, 2026

Search Tags:#workflow #documentation #qa #audit #compliance #review #PHI

Ready to run

npx skills add https://github.com/JustinChaney2023/orate --skill review-qa-workflow

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: review-qa-workflow
Source: https://github.com/JustinChaney2023/orate/tree/main/roles/review-qa-workflow
Command: npx skills add https://github.com/JustinChaney2023/orate --skill review-qa-workflow

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Implement the human review workflow that turns draft OCR/transcription/LLM outputs into approved, exportable clinical documentation with confirmation gates, uncertainty handling, and audit-ready state transitions. Use when building review UI, approval state machine, and QA rules.

Core Features & Use Cases

Review UI & gating: Provide field triage, per-field actions, and evidence-first citations.
State machine & audits: Enforce draft → reviewing → approved → exported with optional rejections and rescan paths, plus detailed audit events.
Use Case: Apply to clinical documentation workflows where PHI must be reviewed before final export and memory embedding.

Quick Start

Design and deploy a review workflow that routes draft outputs through reviewers, required field confirmations, and export gating.

Frequently Asked Questions about review-qa-workflow

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

FAQPage Schema

How do I automate human QA review for clinical documentation containing PHI?▼

Automating human QA review for clinical documentation containing PHI requires a deterministic state machine that routes draft OCR and LLM outputs through reviewing, approved, and exported states with enforced field citations and audit logging. This workflow provides confirmation gates and uncertainty handling to ensure compliance before final export.

What is the best way to implement an approval state machine for draft OCR and LLM transcription outputs?▼

The best way to implement an approval state machine for draft OCR and LLM transcription outputs is to enforce a deterministic progression from draft to reviewing to approved to exported. This approach includes optional rejected or needs_rescan paths, per-field triage actions, and comprehensive audit events to maintain document integrity.

How does a review workflow handle uncertainty and rescan paths for clinical documentation?▼

A review workflow handles uncertainty and rescan paths for clinical documentation by introducing optional states like rejected or needs_rescan within the state machine. This ensures that draft outputs failing field confirmation gates or evidence-first citation requirements are routed back for correction before they can reach the approved and exported stages.

Can I use a deterministic state machine to gate PHI document exports based on field citations?▼

Yes, you can use a deterministic state machine to gate PHI document exports based on field citations. The workflow enforces evidence-first citations and per-field confirmations during the reviewing state, preventing any clinical documentation from transitioning to exported until all gating rules and audit requirements are satisfied.

When do I need audit logging and confirmation gates for clinical documentation QA?▼

You need audit logging and confirmation gates for clinical documentation QA when transitioning PHI-containing draft outputs into export-ready records. Implementing detailed audit events and field-level gating rules ensures that human reviewers verify OCR and transcription accuracy before final memory embedding and export.

Does the clinical documentation QA workflow support export-ready state transitions for LLM outputs?▼

Yes, the clinical documentation QA workflow supports export-ready state transitions for LLM outputs by applying a deterministic state machine. It routes draft content through reviewing and approved states with enforced field citations and audit logging, ensuring LLM-generated text meets compliance standards before final export.