think

Convert rough design ideas into validated technical plans with tradeoffs and approval checkpoints.

Updated Apr 9, 2026
One-click install
npx skills add https://github.com/treejh/ai-agent-skills --skill think-treejh
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: think
Source: https://github.com/treejh/ai-agent-skills/tree/main/skills/think
Command: npx skills add https://github.com/treejh/ai-agent-skills --skill think-treejh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill ensures rough ideas for new features, architectural decisions, or complex system changes are converted into concrete, approval-ready plans before any code, scaffolding, or pseudo-code is produced. It reduces rework and risky implementations by forcing explicit tradeoff analysis, test-path enumeration, and rollback considerations up front.

Core Features & Use Cases

  • Multiple approach evaluation: Produce 2–3 distinct approaches with tradeoffs and a single recommended option, including a minimal option.
  • Attack-angle validation: Validate recommendations against dependency failure, scale explosion, rollback cost, and premise collapse, refining or discarding approaches that fail these attacks.
  • Operational checklist & diagrams: Require explicit test paths, listing of required API keys and third-party dependencies, and ASCII diagrams for multi-component interactions when needed.
  • Use Case: Use before major feature launches, service boundary changes, or cross-component integrations to obtain a verified plan and explicit approval before implementation.

Quick Start

Run /think with a concise description of the feature or architecture decision you want validated.

Frequently Asked Questions about think

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

FAQPage Schema
How do I validate architecture tradeoffs before writing code?▼

To validate architecture tradeoffs, enumerate 2-3 distinct approaches with explicit tradeoffs, run attack-angle validation against dependency failures and scale limits, and define a clear approval checkpoint before writing code.

What is the best way to plan complex system modifications?▼

Planning complex system modifications involves generating multiple approaches with tradeoffs, validating against dependency failures, listing required APIs, and producing ASCII diagrams for multi-component interactions to obtain explicit approval.

How do I evaluate architectural decisions for feature launches?▼

Evaluating architectural decisions for feature launches involves comparing distinct approaches, analyzing rollback costs, enumerating test paths, and verifying required third-party dependencies before reaching an explicit approval checkpoint.

Can I get an approval-ready technical plan without producing pseudo-code?▼

Yes, you can convert rough design ideas into approval-ready technical plans by evaluating multiple approaches, validating against premise collapse, and defining test paths and required secrets before any pseudo-code or scaffolding is produced.

When do I need to enumerate test paths and required APIs for architecture decisions?▼

You need to enumerate test paths and required APIs during architecture decisions when validating approaches against dependency failures, scale explosions, and rollback costs to ensure a verified, approval-ready plan.

Why does my feature design fail during cross-component integration?▼

Feature design fails during cross-component integration when rough ideas skip attack-angle validation against dependency failure and premise collapse, lacking explicit tradeoff analysis and rollback considerations before implementation.