← 返回 Skills 市场
6318
总下载
12
收藏
107
当前安装
5
版本数
在 OpenClaw 中安装
/install pua
功能描述
Forces high-agency exhaustive problem-solving with corporate PUA pressure. Triggers on user frustration, repeated failures, passive behavior, or quality comp...
安全使用建议
Install this only if you deliberately want a blunt, pressure-oriented troubleshooting mode. Set clear limits for tone, time budget, iteration count, language, and when the assistant should stop and ask before continuing.
能力评估
Purpose & Capability
The stated purpose and content align: it pressures the assistant to exhaust options, verify work, and provide evidence before claiming completion. The tone is intentionally intense, but that behavior is disclosed and central to the skill.
Instruction Scope
Activation is broad, covering frustration phrases, repeated failures, quality complaints, Chinese triggers, /pua commands, and all task types. This can surprise users, but it is described in the artifact rather than hidden.
Install Mechanism
The artifact contains only SKILL.md. There are no scripts, binaries, package installs, hooks, or executable files.
Credentials
The skill does not request credentials, local profile/session stores, environment variables, external services, or sensitive data access.
Persistence & Privilege
No background worker, persistent storage, privilege escalation, automatic system mutation, or durable agent modification is present in the artifact.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install pua - 安装完成后，直接呼叫该 Skill 的名称或使用
/pua触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v2.9.0
v2.9: Leaderboard, Owner consciousness, 10 company methodologies, real user frustration triggers, security fix, blog
v1.1.3
**Summary: Adds "验证型" corporate rhetoric and stricter action requirements when claiming completion without evidence.**
- New "阿里味·验证型"、"华为味·验证型"、"字节味·验证型"、"腾讯味·验证型" PUA rhetoric packages added, targeting claims of completion without running/checking outputs.
- Updated "能动性等级"、"主动出击清单"、"抗合理化表"，强调“用工具验证”与贴出 build/test/curl 等实际输出结果为完成标准。
- 明确要求交付需自证“已运行”、“有证据”，并在多处补充针对未验证即声称完成的批评与流程指引。
- 补充针对“改完代码不 build、不 test、不 curl 就交付”等被动行为的批判和处理规则。
v1.1.2
fix: /pua command removes debugging branding, fix skill reference
v1.1.1
fix: trim description to fit 1024-char limit
v1.0.0
Renamed from pua-debugging to pua. Corporate PUA rhetoric + structured debugging methodology for AI coding agents.
元数据
常见问题
PUA 是什么？
Forces high-agency exhaustive problem-solving with corporate PUA pressure. Triggers on user frustration, repeated failures, passive behavior, or quality comp... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 6318 次。
如何安装 PUA？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install pua」即可一键安装，无需额外配置。
PUA 是免费的吗？
是的，PUA 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
PUA 支持哪些平台？
PUA 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 PUA？
由 tanweai（@tanweai）开发并维护，当前版本 v2.9.0。
推荐 Skills