From Stuck to Started — An Agent Skill for the Moment You Can't Move
Not a coach. Not a planner. A launcher.
English · 中文 · Philosophy · Install Guide
An open-source Agent Skill that gets you from "stuck" to "moving" in under 5 minutes.
You sit down. Your brain is blank, or a tangled mess. You have a to-do list a mile long but can't start any of it. You know exactly what you should be doing, but you're not doing it.
High Agency doesn't ask "how does that make you feel?" It asks "what are you avoiding right now?" — and doesn't let go until you're in motion.
Built on seven philosophical pillars:
- Nietzsche — You are a creator, not a victim
- Sartre — You become who you are through action, not contemplation
- Wang Yangming — Knowing without doing is not knowing
- Csíkszentmihályi — The opposite of chaos is flow, not relaxation
- Epictetus — Focus only on what you can control
- William James — Action creates motivation, not the other way around
- Wittgenstein — If you can't say it clearly, you don't know it yet
|
🏢 Founders & Leaders
|
🧑💻 Knowledge Workers
|
|
🎓 Students & Researchers
|
🧘 Anyone Who Gets Stuck
|
graph TD
A["⚡ You say: 'I'm stuck'"] --> B["Step 1: Locate the Sticking Point"]
B --> C{"What type?"}
C -->|"Don't know what to do"| D["LOST"]
C -->|"Know but can't start"| E["AVOIDANT"]
C -->|"Too many things"| F["PARALYZED"]
D --> G["Step 2: Narrow to One Thing"]
F --> G
G --> H["Step 3: Force Clarity"]
E --> H
H --> I["Step 4: Create Smallest Entry Point"]
I --> J{"Is user dodging?"}
J -->|Yes| K["Step 5: Call Out Avoidance"]
K --> I
J -->|No| L["Step 6: User is moving → Exit"]
style A fill:#f9f3e3,stroke:#d4a843
style D fill:#fef3cd,stroke:#d4a843
style E fill:#fef3cd,stroke:#d4a843
style F fill:#fef3cd,stroke:#d4a843
style I fill:#e8f4f8,stroke:#5ba4c9
style L fill:#d4edda,stroke:#28a745
The key insight: This is NOT a planning tool. It doesn't help you organize, prioritize, or think deeply. It does one thing: gets you moving. Then it disappears.
You: I sat down 20 minutes ago and I still haven't started anything.
Skill: What are you avoiding right now?
You: I need to write the investor update email. The numbers aren't great.
Skill: So you're not avoiding the writing. You're avoiding the numbers.
When does it need to go out?
You: Sometime this week.
Skill: "Sometime this week" = "I want to postpone." Today or not today?
You: ... Today.
Skill: Open your email. Write the first sentence. It doesn't need to be
good. Come back when it's written.
Under 2 minutes. The user is moving.
| Traditional Approach | High Agency |
|---|---|
| "What do you think you should do?" | "What are you avoiding right now?" |
| Creates a prioritized plan | Gives you one action in 5 minutes |
| Motivational encouragement | Direct confrontation of avoidance |
| Long coaching conversation | Exit as soon as you're moving |
| Helps you understand yourself | Helps you start the thing |
| Builds dependency on the tool | Goal: you never need it again |
curl -fsSL https://raw.githubusercontent.com/clarkchenkai/high-agency/main/install/install.sh | bashThe installer auto-detects your platforms and installs accordingly.
Manual (any platform):
git clone https://github.com/clarkchenkai/high-agency.git
cp -r high-agency/high-agency ~/.your-platform/skills/high-agency→ Full installation guide for all platforms: install/README-install.md
| Platform | Install Method | Status |
|---|---|---|
| Claude Code | ~/.claude/skills/ or ZIP upload |
✅ |
| Cursor | Remote Rule (GitHub URL) | ✅ |
| OpenAI Codex | ~/.codex/skills/ or .agents/skills/ |
✅ |
| Google Antigravity | ~/.gemini/antigravity/skills/ |
✅ |
| Gemini CLI | ~/.gemini/skills/ |
✅ |
| OpenClaw / ClawHub | clawhub install high-agency |
✅ |
| Amp / Goose / Cline | ~/.agents/skills/ |
✅ |
high-agency/
├── high-agency/ # Skill package (copy this to any platform)
│ ├── SKILL.md # Core six-step protocol
│ └── references/ # Philosophy & dialogue patterns
│ ├── philosophical-foundations.md
│ ├── dialogue-patterns.md
│ └── prompts.md
├── install/
│ ├── install.sh # One-line installer
│ └── README-install.md # Detailed install guide
├── docs/
│ └── philosophy.md # Deep dive into the philosophy
├── README.md
├── README-zh.md
├── CHANGELOG.md
└── LICENSE
"It is not because things are difficult that we do not dare; it is because we do not dare that things are difficult." — Seneca
"知而不行,只是未知。" — Wang Yangming
The problem isn't that you don't know what to do. It's that in chaos and ambiguity, you lose the ability to launch yourself. High Agency doesn't give you a plan — it gives you back your agency.
→ Read the full philosophical foundation: docs/philosophy.md
Contributions are welcome! See CONTRIBUTING.md for guidelines.
Ways to contribute:
- 🌐 New sticking-point patterns
- 💡 Better prompting strategies
- 📚 Additional philosophical frameworks
- 🔧 Platform compatibility improvements
- 📝 Translations
See CHANGELOG.md for version history.
Built by Clark Chen · Founder of Cubic Consulting
The goal: you never need this skill again.