instruct

Convert human explanations into unambiguous AI-ready instruction recipes with atomic steps and flagged assumptions.

Updated Apr 28, 2026
One-click install
npx skills add https://github.com/ChazzCoin/claude-kit --skill instruct
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: instruct
Source: https://github.com/ChazzCoin/claude-kit/tree/main/kit/skills/instruct
Command: npx skills add https://github.com/ChazzCoin/claude-kit --skill instruct

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Converts a human explanation of how to do something into an unambiguous AI-formatted instruction recipe. It decomposes the input into the smallest sensible deliverable steps, performs a mock run-through to catch missing or out-of-order steps, flags every assumption inline, and returns a formatted recipe that an AI (or human) can follow without guessing. It is a pre-task: the deliverable is the instruction set and it does not execute the steps.

Core Features & Use Cases

  • Decomposes vague instructions into atomic, verifiable steps with a fixed recipe shape.
  • Performs a mock run-through to surface gaps and assumptions, recording them for traceability.
  • Produces a ready-to-use instruction recipe that guides subsequent AI sessions or human operators without executing the steps.

Quick Start

Paste messy instructions and run /instruct to receive a clean, atomic step recipe.

Frequently Asked Questions about instruct

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

FAQPage Schema
How do I turn messy notes into clear AI instructions?

To turn messy notes into clear AI instructions, paste your rough task lists or brain-dumps into the chat. The tool decomposes the prose into atomic, verifiable steps and returns a formatted, unambiguous instruction recipe.

What is the best way to format vague task lists for AI processing?

The best way to format vague task lists is to decompose them into atomic steps with a fixed recipe shape. This process flags inline assumptions and includes a mock run-through to catch missing or out-of-order steps before the AI executes the task.

How do I flag assumptions and gaps in my AI prompt instructions?

To flag assumptions and gaps in AI prompt instructions, the system performs a mock run-through of your decomposed steps. This dry-run surfaces missing dependencies and records inline assumptions directly within the generated instruction recipe for traceability.

Does this instruction recipe tool execute the steps it generates?

No, the instruction recipe tool does not execute the steps. It acts strictly as a pre-task that produces a chat-only, non-executing formatted recipe, ensuring an AI or human operator can follow the atomic steps without guessing.

Can I use brain-dumps and prose to generate step-by-step AI recipes?

Yes, you can use brain-dumps and prose to generate step-by-step AI recipes. The tool accepts unstructured human explanations and converts them into a structured, atomic recipe format ready to guide subsequent AI work.

Why do my AI prompts fail when I provide rough task lists?

AI prompts often fail with rough task lists because they lack atomic, verifiable steps. Converting your notes into a structured instruction recipe with a mock run-through prevents guessing by surfacing gaps and flagging inline assumptions.