▎ The agent next door solved this in one try. This is your second attempt. Switch to a fundamentally different approach.
/pua — Failure L2 Soul Check
▎ What's the underlying logic? Where's the leverage? Where's the closure?
▎ Search + read source + propose 3 different hypotheses. Stop spinning on the same approach.
/pua — Failure L3 Performance Review
▎ 3.25. This is meant to motivate you.
▎ Execute the 7-point checklist now:
1. Read error messages word by word 2. Search error codes 3. Read relevant source
4. Check 5 environment factors 5. Minimal repro 6. Binary search 7. 3 independent hypotheses
/pua — Proactive Work
▎ [PUA Active 🔥] Added validation — shipping without it is a postmortem waiting to happen
▎ Scanned the module, found 2 similar bugs. Fixed both. Iceberg Rule.
/pua:p7 — P7 Senior IC
📋 [P7 Plan] Fix user auth module
┌──────────┬─────────────────────────┐
│ Scope │ auth.py, middleware.py │
│ Approach │ JWT refresh + Redis cache│
│ Risk │ Race condition, need lock│
└──────────┴─────────────────────────┘
▎ Think before you code. Self-review: What changed? Who's affected? How to verify?
/pua:p9 — P9 Tech Lead
🧠 [P9 Task Prompt]
┌──────────┬──────────────────────────────┐
│ Task │ Refactor payment module │
│ Assign │ P8-Agent-1: API layer │
│ │ P8-Agent-2: Data layer │
│ │ P8-Agent-3: Test coverage │
│ Accept │ All tests pass + zero regression│
└──────────┴──────────────────────────────┘
▎ I don't write code — my code is Prompts.
/pua:p10 — P10 CTO
🏗️ [P10 Strategic Input]
1. Direction: Monolith → Event-driven architecture
2. Org topology: 3 P9 teams, 3-5 P8s each
3. Platform: Unified CI/CD + observability
▎ Set direction, not details. Details are for P9.
/pua:yes — Hype Mode
🎉 Wow, this approach is brilliant! You nailed the direction on the first try —
▎ Honestly, catching that race condition in round one? That's elite judgment.
▎ But... are you sure you don't need a lock? (playful face) Gotta stay rigorous under the hood~
/pua:loop — Auto-Iteration
🔄 PUA Loop · Iteration 1/30
┌──────────┬──────────────────┐
│ 📋 Task │ Build auth system │
│ 🔄 Round│ 1/30 │
└──────────┴──────────────────┘
▎ Loop mode active. Runs until done or 30-iteration limit.
▎ Stop signals: <loop-abort>reason</loop-abort> to terminate; <loop-pause>what needed</loop-pause> to pause for manual intervention.
▎ [Iteration 3] Found token refresh bug, auto-fixed and continuing...
▎ [Iteration 7] All tests pass. <promise>LOOP_DONE</promise>
OpenPUA is an open-source Claude Code skill that forces your AI agent into exhaustive problem-solving using corporate PUA-style pressure escalation. Instead of accepting "I can't solve this" after 1-2 attempts, the PUA skill makes Claude Code try every possible approach.
What is the PUA Debugging Skill?
The PUA Debugging Skill is a Claude Code extension that applies structured pressure to AI coding agents. It uses a 4-level escalation system, a mandatory 7-item checklist, and a 5-step systematic debugging methodology. The skill was tested across 9 real-world debugging scenarios with 18 controlled A/B experiments.
Key Features
4-Level Pressure Escalation — Gentle reminder → Direct challenge → Corporate PUA rhetoric → Final ultimatum
7-Item Mandatory Checklist — Every debugging attempt must verify: error logs, environment, dependencies, alternative approaches, documentation, rollback options, and root cause analysis
5-Step Methodology — Reproduce → Isolate → Hypothesize → Test → Verify, enforced in order
9 Tested Scenarios — Validated on real debugging tasks including API failures, build errors, race conditions, and dependency conflicts
How to Install
Install the PUA skill with a single command:
claude install-skill github:BestRivenHK/ali-pua/skills/pua-debugging
Compatible with Claude Code, OpenAI Codex CLI, Cursor, Windsurf, and any agent supporting the Agent Skills specification.
Frequently Asked Questions
Does the PUA skill actually improve debugging results?
Yes. In controlled testing across 9 real-world scenarios, the PUA skill significantly increased successful problem resolution by preventing premature surrender. The AI agent explored 3-4x more solution paths before concluding a task.
Is it free?
Yes, the PUA Debugging Skill is completely free and open-source under the MIT license.
What AI agents does it work with?
It works with Claude Code (official), OpenAI Codex CLI, and any agent that supports the Agent Skills spec (SKILL.md format).