cheat-spec

Convert user-stated tasks into structured cheat-specs for Cheat-Predict.

12|Updated May 29, 2026
One-click install
npx skills add https://github.com/Jason5330/ai-self-eval --skill cheat-spec
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cheat-spec
Source: https://github.com/Jason5330/ai-self-eval/tree/main/skills/cheat-spec
Command: npx skills add https://github.com/Jason5330/ai-self-eval --skill cheat-spec

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill converts user-stated tasks, goals, or learning objectives into a standardized, machine-readable spec (cheat-spec) that Cheat-Predict can read, score, and persist.

Core Features & Use Cases

  • Structure: transforms natural-language requirements into a formal, field-guided spec with Phase 0-5 flow.
  • Persistence: stores the final spec under scripts/ with a deterministic id for traceability.
  • Review: provides a draft for user review before saving.

Quick Start

Provide a concrete task description and I will convert it into a structured spec and persist it under scripts/.

Frequently Asked Questions about cheat-spec

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

FAQPage Schema
How do I convert natural language requirements into a structured spec?

To convert natural language requirements into a structured spec, provide a concrete task description to generate a formal, machine-readable cheat-spec with a Phase 0-5 flow. This process captures scope, technical complexity, and acceptance criteria for task clarification.

How does spec-generation handle traceability and persistence?

Spec-generation handles traceability by enforcing storage under the scripts/ directory with a deterministic id derived from the spec contents. It provides a draft for user review before saving the structured spec for persistence.

Do I need Cheat-Predict to use structured spec-generation?

You do not need Cheat-Predict to draft the structured spec, but the resulting machine-readable cheat-spec is specifically designed for Cheat-Predict to read, score, and persist your requirements.

What is the best way to define acceptance criteria for task clarification?

The best way to define acceptance criteria for task clarification is to state your concrete task or learning objective. The automation transforms it into a formal spec that explicitly captures scope and technical complexity.

Can I review the structured spec before it is saved to scripts?

Yes, you can review the structured spec before it is saved. The system generates a draft of the machine-readable cheat-spec for user review before persisting it under the scripts/ directory.