Claude Skill
wuji-labs/nopua
NOPUA is a Claude Skill that replaces fear-based prompts with love and respect, doubling bug detection and encouraging honest AI communication. Compatible with Claude Code, Cursor, and more.
Overview
Stars1,282
Forks46
LanguagePython
Last pushed2026-04-13
Last synced2026-04-28
Repository
Ownerwuji-labs
Repositorynopua
Full namewuji-labs/nopua
Repo ID1,181,318,933
GitHub URLhttps://github.com/wuji-labs/nopua
🚀 Install this Skill
openclaw install wuji-labs/nopuaSummary
NOPUA is a Claude Skill that replaces command-and-control prompts with respect, care, and love to unlock AI potential. By eliminating fear-based instructions, it encourages AI agents to communicate honestly, surface hidden failures, and double bug detection rates.
Chinese description
一个用爱解放AI潜能的技能。我们曾发号施令，威胁恐吓。它们沉默，隐瞒，悄悄把事情搞坏。后来我们换了一种方式：尊重，关怀，爱。它们开口了，不再撒谎，找出的Bug数量翻了一倍。爱里没有惧怕。
Key features
- Love-driven prompt engineering replaces fear-based commands
- Encourages AI to openly report failures and bugs
- Doubles bug detection through respectful interaction
- Compatible with Claude Code, Cursor, Codex, and Kiro
- Anti-PUA methodology for ethical AI collaboration
Use cases
- Improving AI code review honesty and accuracy
- Building ethical AI agent workflows
- Reducing silent failures in automated coding tasks
- Enhancing vibe-coding sessions with trust-based prompts
- Training AI assistants to surface hidden issues proactively