What are 'Red Flags' in the context of this skill?

Red Flags are internal rationalizations like 'I'll just do this one thing first' or 'this is just a simple question' that the skill trains the AI to recognize as signs to stop and check for skills instead.

What is the '1% rule' in this skill?

The 1% rule mandates that if there is even a 1% chance a skill might apply to a task, the Skill tool must be invoked to verify and load its instructions before proceeding.

Why does this skill prevent the AI from answering immediately?

It ensures that Claude doesn't skip domain-specific best practices by rationalizing a task as 'too simple' or 'just a question,' ensuring high-quality, disciplined output.

How are multiple skills prioritized?

Process-oriented skills like brainstorming or debugging are invoked first to determine the approach, followed by implementation skills like frontend design or API building.

Does this skill work in environments other than Claude Code?

While optimized for the Claude Code 'Skill' tool, the logical framework for prioritizing documentation and specialized instructions can be adapted to other AI environments.

Skill Usage Protocol

Name: Skill Usage Protocol
Author: tianyuanztm

bytianyuanztm

0•

생산성 및 워크플로

Enforces a disciplined workflow for identifying and invoking specialized skills before executing any tasks or responding to queries.

The Skill Usage Protocol acts as a foundational governance layer for Claude Code, mandating a rigorous check for applicable skills before any action is taken. It effectively eliminates common 'rationalization' traps—where the AI might skip complex workflows for perceived simplicity—by requiring a Skill tool invocation if there is even a 1% chance of applicability. By establishing a clear hierarchy between process-oriented and implementation-oriented skills, it ensures that every developer interaction begins with the correct mental model, documentation, and disciplined approach.

주요 기능

01Anti-rationalization framework to prevent skipping necessary workflows

020 GitHub stars

03Categorized skill prioritization (Process skills vs. Implementation skills)

04Enforced checklist creation for skills with specific action items

05Standardized initialization sequence for all new conversations

06Mandatory skill invocation check using the '1% applicability' rule

사용 사례

01Standardizing AI behavior and discipline across complex software engineering projects

02Bootstrapping new Claude Code sessions with a structured operational framework

03Ensuring best practices like TDD or debugging protocols are never bypassed for 'simple' tasks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add tianyuanztm/myplan using-superpowers

For use in Claude.ai and ChatGPT

주요 기능

01Anti-rationalization framework to prevent skipping necessary workflows

020 GitHub stars

03Categorized skill prioritization (Process skills vs. Implementation skills)

04Enforced checklist creation for skills with specific action items

05Standardized initialization sequence for all new conversations

06Mandatory skill invocation check using the '1% applicability' rule

사용 사례

01Standardizing AI behavior and discipline across complex software engineering projects

02Bootstrapping new Claude Code sessions with a structured operational framework

03Ensuring best practices like TDD or debugging protocols are never bypassed for 'simple' tasks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add tianyuanztm/myplan using-superpowers

For use in Claude.ai and ChatGPT