narrative-judge

Assess rewritten narrative beats against insert and original for contradiction resolution.

178|18|Updated May 3, 2026
One-click install
npx skills add https://github.com/UfukNode/Noustiny --skill narrative-judge
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: narrative-judge
Source: https://github.com/UfukNode/Noustiny/tree/main/hermes-additions/skills/creative/narrative-judge
Command: npx skills add https://github.com/UfukNode/Noustiny --skill narrative-judge

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Adjudicates a rewritten narrative beat against the insert-trigger and the original beat, returning a strict JSON verdict to preserve continuity.

Core Features & Use Cases

  • Final-approval gate for reharmonize passes, ensuring rewrites resolve contradictions without introducing new ones.
  • Evaluates the rewritten beat against the original and the trigger, providing an approved flag, a score from 0 to 1, and a one-sentence justification.
  • Works within Hermes narrative pipelines to gate canonization of rewrites and maintain dramaturgical integrity.

Quick Start

Tell the AI to decide whether the latest rewrite resolves the insert's contradiction while preserving the original beat's purpose.

Frequently Asked Questions about narrative-judge

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

FAQPage Schema
How do I validate narrative continuity after rewriting a story beat?

To validate narrative continuity after rewriting a story beat, you need an adjudication step that evaluates the rewritten beat against the original and the insert-trigger. This process checks whether the rewrite resolves contradictions while preserving the original beat's dramaturgical function.

What is a reharmonize workflow in interactive fiction?

A reharmonize workflow in interactive fiction is a narrative pipeline process that resolves contradictions introduced by new story inserts. It uses a final-approval gate to ensure rewritten beats maintain canonization and dramaturgical integrity without introducing new continuity errors.

How do I get a structured JSON verdict for automated narrative approvals?

You can get a structured JSON verdict for automated narrative approvals by evaluating insert, original, and rewritten beats in a single pass. The output contains an approved boolean, a numeric score between 0 and 1, and a one-sentence justification reason.

Can I use this narrative adjudication approach within Hermes pipelines?

Yes, you can use this narrative adjudication approach within Hermes pipelines. It functions specifically as a final-approval gate to evaluate rewrites, ensuring that contradictions are fully resolved before canonization occurs within the narrative pipeline.

What are the limitations of automated narrative rewrite evaluation?

A limitation of automated narrative rewrite evaluation is that it provides a strict binary approval flag and a single numeric score, which may not capture nuanced dramaturgical contexts. It strictly evaluates whether the insert's contradiction is resolved without introducing new ones.