dragosroua/claude-assess-decide-do-mega-prompt
Developer tools이 페이지는 AI로 자동 번역되었습니다. 기술적인 세부 사항은 원문을 확인하세요. 원문 보기
First human cognitive framework integrated into an LLM. Claude learns to detect if you're assessing, deciding, or doing—and responds appropriately.
개요
This is the first integration of a human cognitive framework into a Large Language Model. Not training AI to be smarter—training AI to recognize human mental states. The result: Cognitive alignment that users describe as "weirdly empathic." What started as a megaprompt is now a modular ecosystem: Start here. Grab the skills. Try the Cowork plugin. AI assistants create subtle but constant friction: - → AI jumps to solutions - → AI suggests rethinking everything - → AI provides complete answers AI doesn't understand in your thinking process. The teaches Claude to recognize three fundamental cognitive states: - - Exploring possibilities without pressure to commit - - Choosing direction and allocating resources - - Executing and completing with focus
README
원문 README의 일부입니다. 원문 보기.
When AI Understands Where You Are in Your Thinking
What if AI didn’t just answer questions, but understood whether you’re exploring, deciding, or executing?
This is the first integration of a human cognitive framework into a Large Language Model. Not training AI to be smarter—training AI to recognize human mental states.
The result: Cognitive alignment that users describe as “weirdly empathic.”
🚀 This Project Has Evolved
What started as a megaprompt is now a modular ecosystem:
Repository Purpose This Repo Core framework, documentation, philosophy ADD Skills Modular Claude Code skills (94% less context overhead) ADD Cowork Plugin Full ADD integration for Claude Cowork — no coding required New to ADD? Start here. Using Claude Code? Grab the skills. Not a coder? Try the Cowork plugin.
The Problem
AI assistants create subtle but constant friction:
- You’re exploring options → AI jumps to solutions
- You’re mid-execution → AI suggests rethinking everything
- You need support → AI provides complete answers
AI doesn’t understand where you are in your thinking process.
The Solution
The Assess-Decide-Do (ADD) framework teaches Claude to recognize three fundamental cognitive states:
- 🔴 Assess - Exploring possibilities without pressure to commit
- 🟠 Decide - Choosing direction and allocating resources
- 🟢 Do - Executing and completing with focus
When Claude understands your realm, it provides cognitive support instead of generic responses.
Quick Start → | Why This Matters →
What You’ll Experience
Cognitive Benefits
✅ Reduced mental friction - Responses match your actual needs
✅ Clearer thinking - Realm boundaries reduce overwhelm
✅ Natural flow - Smooth progression through work phases
✅ ADHD-friendly - Cognitive load management by design
Workflow Benefits
✅ Fewer misaligned responses - Claude detects your mental state
✅ Smoother execution - No premature solutions or re-evaluation
✅ Better completion rates - Support when and how you need it
✅ Balanced rhythm - Not just perpetual task-churning
The Relational Shift (Unexpected)
추천 도구
다른 키워드를 입력하거나 필터를 제거해 보세요.
설치
npx skillfish add dragosroua/claude-assess-decide-do-mega-prompt