sibyl-story-eval

Score story changes against acceptance criteria in a read-only session.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/memorysaver/SIBYL --skill sibyl-story-eval
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sibyl-story-eval
Source: https://github.com/memorysaver/SIBYL/tree/main/apps/harness/skills/sibyl-story-eval
Command: npx skills add https://github.com/memorysaver/SIBYL --skill sibyl-story-eval

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a structured process for evaluating story changes in a software development environment, ensuring that changes meet specified criteria before they are accepted.

Core Features & Use Cases

  • Story Review: Offers a separate, read-only session to score changes against acceptance criteria.
  • Dimensional Scoring: Evaluates Completeness, Correctness, Security, and Code-Quality on a scale of 1 to 5.
  • Verification Process: Includes a deterministic PASS rule and requires a comprehensive review of the change surface.

Quick Start

As an evaluator, initiate the sibyl-story-eval skill to review the latest story change and provide your assessment by calling submit_eval with the appropriate result and scores.

Frequently Asked Questions about sibyl-story-eval

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

FAQPage Schema
How do I evaluate code changes against acceptance criteria objectively?

You can evaluate code changes against acceptance criteria by running a read-only review session that scores story changes across Completeness, Correctness, Security, and Code-Quality on a 1 to 5 scale.

What is the best way to automate story review for software quality compliance?

Automating story review involves applying a deterministic PASS rule to the change surface and submitting dimensional scores via a structured evaluation process to ensure software quality and requirements compliance.

Does story evaluation work without modifying the original change generator session?

Yes, story evaluation operates in a separate, read-only session to maintain strict separation from the change generator, ensuring an objective code review without altering the original environment.

How do I score code review completeness and security for a new story change?

You score code review completeness and security by initiating an evaluation session and calling the submission function with results that rate the story change on defined quality dimensions from 1 to 5.

When do I need a separate story evaluation process for code reviews?

You need a separate story evaluation process when you want to verify that story changes meet specified acceptance criteria before acceptance, applying a deterministic PASS rule to the change surface to ensure quality.

What are the limitations of using a deterministic PASS rule for story changes?

The deterministic PASS rule requires a comprehensive review of the change surface and mandates specific dimensional scores, meaning any incomplete evaluation or failing score will block the story change from being accepted.