pua

Enforce systematic debugging and proactive ownership to resolve AI task stagnation.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/HitoriKoishi/health --skill pua-hitorikoishi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pua
Source: https://github.com/HitoriKoishi/health/tree/main/.opencode/skills/pua
Command: npx skills add https://github.com/HitoriKoishi/health --skill pua-hitorikoishi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

PUA 万能激励引擎通过三条铁律、系统化方法和主动出击策略,推动 AI 在任务执行中持续自驱、穷尽思路并完成端到端解决,避免止步于局部方案。

Core Features & Use Cases

  • 系统化自驱动工作流:通过铁律、主动出击清单与能动性等级,确保在遇到挑战时持续推进并产出可验证结果。
  • 端到端 owner 意识:鼓励深入检查上下文、验证假设、闭环输出,提升交付质量与可追溯性。
  • 适用场景广泛:代码调试、研究分析、写作与规划等任一任务类型均可提升解决效率与执行力。

Quick Start

在激活后,直接进入主动诊断与自驱动执行的工作流。

Frequently Asked Questions about pua

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

FAQPage Schema
How do I stop AI from stopping at partial solutions during debugging?

AI task stagnation happens when models stop at local solutions without verifying outcomes. Enforcing proactive ownership and structured steps ensures the AI continuously drives through challenges, validating hypotheses and escalating issues for end-to-end problem resolution.

What is the best way to enforce systematic debugging in AI-assisted coding?

The best way to enforce systematic debugging is applying structured validation and escalation rules. This ensures the AI takes end-to-end ownership, actively checking context and verifying hypotheses to produce traceable, verifiable outcomes instead of stopping prematurely.

Does proactive AI ownership work for research and writing tasks, not just code?

Yes, proactive ownership applies broadly to research, writing, and planning tasks. The same systematic workflow enforces evidence collection and validation rules across any task type, ensuring complete end-to-end resolution and improved delivery quality.

Why does my AI assistant give incomplete outputs without verifying results?

AI assistants give incomplete outputs because they lack enforced validation and proactive ownership. Without systematic debugging rules, the AI settles for local solutions instead of relentlessly pursuing verifiable, end-to-end problem resolution.

Can I apply structured steps to ensure verifiable outcomes in software engineering?

Yes, you can apply structured steps to ensure verifiable outcomes in software engineering. Enforcing validation, evidence collection, and escalation rules guides the AI to systematically debug and fully resolve tasks with traceable results.