scoring-checks

Implement TypeScript scoring checks with constants and weights summing to 100.

1.2k|119|Updated Mar 10, 2026
One-click install
npx skills add https://github.com/caliber-ai-org/ai-setup --skill scoring-checks
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: scoring-checks
Source: https://github.com/caliber-ai-org/ai-setup/tree/main/.cursor/skills/scoring-checks
Command: npx skills add https://github.com/caliber-ai-org/ai-setup --skill scoring-checks

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Streamlines the creation of deterministic, pluggable scoring checks that evaluate project quality without randomness, enabling consistent scoring results across builds.

Core Features & Use Cases

  • Deterministic Check[] generators that return a score and reasons for each check
  • Seamless integration with the existing scoring index and weight system to maintain a 100-point total
  • Support for standard export patterns and TypeScript typing to ease onboarding and testing

Quick Start

Create a new check in src/scoring/checks, export it from the index, and wire it into the main score() function.

Frequently Asked Questions about scoring-checks

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

FAQPage Schema
How do I add deterministic scoring checks to a TypeScript pipeline?

To add deterministic scoring checks, implement a TypeScript function returning a Check[] array using constants from src/scoring/constants.ts, export it, and wire it into the main score() function. You must also update the check weights so they consistently sum to 100.

What is a deterministic scoring check in software architecture?

A deterministic scoring check is a pluggable TypeScript function that evaluates project quality and returns a score with reasons without randomness. This ensures stable and reproducible assessment results across multiple pipeline builds.

How do I integrate a new check into an existing scoring index?

You integrate a new check by creating it in src/scoring/checks, exporting it from the index, and wiring it into the main score() function. You must then adjust the weight system to maintain the required 100-point total across all checks.

Do I need to update weights when adding new scoring checks?

Yes, you need to update weights when adding new scoring checks. The scoring system requires you to adjust the weights for all checks so that they continue to sum to exactly 100 points within the pipeline.

Can I use standard TypeScript typing for custom scoring checks?

Yes, you can use standard TypeScript typing for custom scoring checks. The system supports standard export patterns and TypeScript typing to ease onboarding and testing for deterministic Check[] generators.