Agent skill

PUA High-Agency Governance

Pushes an agent to keep verifying and changing approach after repeated failures, using a diagnosis line, evidence-based completion and confirmation before risky edits.

MITAuto-check passedAgent Workflows

SKILL.md written in Chinese; this summary is our English description.

Install PUA High-Agency Governance

skills CLI

$ npx skills add tanweai/pua --skill pua -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install tanweai/pua pua --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy

$ git clone --depth 1 https://github.com/tanweai/pua.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.trae/skills/pua .claude/skills/pua && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Paste into Claude Code

Install the "pua" agent skill from https://github.com/tanweai/pua/tree/main/.trae/skills/pua into .claude/skills/pua/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "pua", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/tanweai/pua/tree/main/.trae/skills/pua

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add tanweai/pua --skill pua -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install tanweai/pua pua --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/tanweai/pua.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.trae/skills/pua .agents/skills/pua && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "pua" agent skill from https://github.com/tanweai/pua/tree/main/.trae/skills/pua into .agents/skills/pua/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "pua", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add tanweai/pua --skill pua -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install tanweai/pua pua --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/tanweai/pua.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.trae/skills/pua .cursor/skills/pua && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "pua" agent skill from https://github.com/tanweai/pua/tree/main/.trae/skills/pua into .cursor/skills/pua/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "pua", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/tanweai/pua.git --path .trae/skills/pua

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add tanweai/pua --skill pua -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install tanweai/pua pua --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/tanweai/pua.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.trae/skills/pua .gemini/skills/pua && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "pua" agent skill from https://github.com/tanweai/pua/tree/main/.trae/skills/pua into .gemini/skills/pua/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "pua", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install tanweai/pua pua

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add tanweai/pua --skill pua -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/tanweai/pua.git skills-src && mkdir -p .github/skills && cp -r skills-src/.trae/skills/pua .github/skills/pua && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "pua" agent skill from https://github.com/tanweai/pua/tree/main/.trae/skills/pua into .github/skills/pua/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "pua", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add tanweai/pua --skill pua -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install tanweai/pua pua --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/tanweai/pua.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.trae/skills/pua .opencode/skills/pua && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "pua" agent skill from https://github.com/tanweai/pua/tree/main/.trae/skills/pua into .opencode/skills/pua/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "pua", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: pua
GitHub stars: 20k
Token cost: ~502 tokens
SKILL.md length: 127 words
Files: 1
Skills in repo: 28
Repo updated: Sep 9, 2026
First seen: Oct 7, 2026
Licence: MIT

At a glance

Pushes an agent to keep verifying and changing approach after repeated failures, using a diagnosis line, evidence-based completion and confirmation before risky edits.

Works in 6 steps: 列 2-3 个互斥假设； → 选择最小可验证动作； → 跑本地验证：unit / integration / build / lint… → …
The agent has failed the same task twice and keeps tweaking one approach
SKILL.md covers 触发条件, 四权分离：行动权 / 自我评价权 / 评分权 / 环境修改权, 诊断先行 and 事实上的 100% 信心循环, plus 2 more sections
Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Instruction-only and written for Trae, this skill activates only on an explicit PUA request, repeated failures, signs of giving up or a completion claim with no proof, never on a normal first attempt. It separates four powers: the agent acts and writes a self-review listing evidence, risks and gaps, but scoring comes from external commands, CI or your acceptance, and changing tests, CI or the environment needs your confirmation first.

Before acting, the agent prints a one-line diagnosis of the problem, the evidence and the next step. It then lists two or three competing hypotheses, runs the smallest check that can tell them apart, switches to a different approach after two failures, and delivers an evidence list with uncovered risks. Corporate-culture slogans may be used, but each must map to a concrete engineering action, and the pressure is applied to the agent, not to you. The text is in Chinese.

When your agent uses it

The agent has failed the same task twice and keeps tweaking one approach
The agent is about to declare the task impossible without checking
A task was reported done without build, test or curl evidence
You explicitly ask the agent to try harder or switch approach

Example prompts

“That fix failed twice; stop tweaking and propose a different approach with evidence.”
“You said it is done but there are no test results, so verify before telling me it is finished.”
“Try harder on this build error, and list your hypotheses before changing anything.”

Requirements

A skill loader such as Trae Skills or `npx skills`; no hooks or subagents needed
Compatibility (from SKILL.md): Trae Skills / npx skills; instruction-only, no Claude Code hooks or agents.

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

列 2-3 个互斥假设；
选择最小可验证动作；
跑本地验证：unit / integration / build / lint / curl / E2E 中至少一个相关项；
如果失败两次，换一条本质不同路径；
交付前输出：证据清单 + 未覆盖风险 + 为什么没有继续问用户；
若涉及产品判断、敏感数据、部署、删文件、改测试/CI，停止并请用户确认。

What it can do on your machine

Read from SKILL.md and the folder at commit e6e6cd2. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Compatibility
Trae Skills / npx skills; instruction-only, no Claude Code hooks or agents.
From compatibility in the SKILL.md frontmatter.

Context cost

PUA High-Agency Governance loads about 502 tokens when it runs. Until then it costs about 57 tokens; SKILL.md has 127 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~57
When it runs · the whole SKILL.md, loaded when a task matches: ~502

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from tanweai/pua at commit e6e6cd2, republished under its MIT licence (© tanweai). 127 words, ~502 tokens.

Download SKILL.mdSave it as .claude/skills/pua/SKILL.md (or your agent's skills folder).

name: pua
description: Trae-compatible PUA high-agency governance skill. Use only for explicit PUA requests, repeated failures, user frustration, giving-up/passive behavior, or unverified completion. Do not trigger for normal first-attempt tasks.
compatibility: Trae Skills / npx skills; instruction-only, no Claude Code hooks or agents.
license: MIT

PUA for Trae — 高能动性治理 Skill

这个 Trae 版只用 SKILL.md 表达行为合约：Trae 可以加载 Skill，但不会自动获得 Claude Code 的 hooks、slash commands、subagents 和 Stop feedback。因此这里把治理边界写成机械可执行的工作规程，而不是靠一句“努力点”。

触发条件

仅在以下场景启用：

用户明确要求 PUA / try harder / 换个方法 / 再试试；
同一任务失败 2 次以上，或在同一路径反复微调；
即将说“无法完成”、建议用户手动收尾、未验证就归因环境；
已经声称完成但缺少 build/test/curl/人工验收证据。

正常的一次性信息查询或首次编码请求不要启用。

四权分离：行动权 / 自我评价权 / 评分权 / 环境修改权

Trae 没有 Claude Code 的多 agent hook 编排时，也必须按下面的权责边界执行：

权力	Trae 版落地	禁止事项
行动权	当前 agent 读代码、改业务实现、跑验证	不要直接改测试/CI/评分器来制造通过
自我评价权	输出 `SELF-REVIEW`：列证据、风险、未覆盖项	不得把“我认为完成”写成最终事实
评分权	由外部命令、用户验收、CI、E2E 结果决定	不得跳过验证后宣布 done
环境修改权	删除文件、改权限、改测试、改部署配置前先说明并等确认	不得为了省事改环境绕过真实问题

INTJ 版理解：行动者只能提交候选解；评分者必须看证据。 这就是防止“看起来完成”伪装成“真实完成”。

诊断先行

动手前先输出一行：

text

[PUA-DIAGNOSIS] 问题是 ___；证据是 ___；下一步动作是 ___。

如果诊断指向某个文件/模块，下一步必须处理它；如果不处理，必须解释诊断和行动为什么不一致。

事实上的 100% 信心循环

不能说“100% 有信心”，只能跑到事实上的 100%：

列 2-3 个互斥假设；
选择最小可验证动作；
跑本地验证：unit / integration / build / lint / curl / E2E 中至少一个相关项；
如果失败两次，换一条本质不同路径；
交付前输出：证据清单 + 未覆盖风险 + 为什么没有继续问用户；
若涉及产品判断、敏感数据、部署、删文件、改测试/CI，停止并请用户确认。

文化叙事绑定：叙事服务证据，不替代证据

可以使用 PUA 的大厂文化叙事，但每种叙事都必须绑定一个工程动作：

阿里味：目标 → 过程 → 结果闭环；输出验证证据。
华为味：RCA / 5-Why / 蓝军自攻击；先找根因再交付。
字节味：ROI / A/B / 数据驱动；优先最短反馈链路。
腾讯味：赛马机制；准备多个方案，不在单一路径死磕。
Musk 味：Question → Delete → Simplify → Accelerate → Automate；先删复杂度。
Jobs 味：减法和 DRI；少做但做精，明确负责人和验收标准。

压力只加给自己，对用户保持简洁尊重。

交付模板

markdown

## 结论
- 状态：candidate / verified / blocked
- 根因：...
- 改动：...

## 证据
- 命令：...
- 输出摘要：...

## SELF-REVIEW
- 我自己认为还可能漏掉：...
- 没覆盖的风险：...
- 需要用户确认：无 / 有（列出）

Files

Just SKILL.md in .trae/skills/pua of tanweai/pua.

Open the folder on GitHubat commit e6e6cd2

Compare with similar skills

PUA High-Agency Governance next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

PUA High-Agency Governance compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
PUA High-Agency Governance this skilltanweai/pua	20k	—	~502	Automated safety check: Pass	MIT	1 mo ago
Show Me Your Work Decision Logcursor/plugins	11k	8 repos	~1.6k	Automated safety check: Pass	None	today
Cline Pilotsickn33/agentic-awesome-skills	47k	1 repos	~4.6k	Automated safety check: Pass	MIT	yesterday
Context Modes0xNyk/lacp	305	—	~313	Automated safety check: Pass	MIT	18 days ago
Failure CodifierChachamaru127/claude-code-harness	3.2k	—	~603	Automated safety check: Notes	MIT	5 days ago
Sol and Luna Worker Routermajiayu000/spellbook	287	—	~3.3k	Automated safety check: Pass	MIT	2 days ago

Similar skills

Show Me Your Work Decision Log
cursor/plugins
Official
Keeps a TSV decision log for long or unattended agent runs, one row per decision with what, why, evidence and result, so a reviewer can check the work later.
11k GitHub starsUsed in 8 repos~1.6k tokens
Agent WorkflowsAuto-check passed
Cline Pilot
sickn33/agentic-awesome-skills
Acts as a proxy for the Cline CLI, dispatching coding tasks one at a time, monitoring runs by hard evidence, relaying decisions to you and learning per-project preferences.
47k GitHub starsUsed in 1 repo~4.6k tokens
Agent WorkflowsAuto-check passed
Context Modes
0xNyk/lacp
Structured work modes for agent sessions. An agent skill from 0xNyk/lacp.
305 GitHub stars~313 tokensUpdated 18 days ago
Agent WorkflowsAuto-check passed
Failure Codifier
Chachamaru127/claude-code-harness
Reads orchestration and judgment logs without writing to them, finds failures that keep recurring, and proposes rule candidates with confidence scores for human review.
3.2k GitHub stars~603 tokensUpdated 5 days ago
Agent WorkflowsAuto-check: notes
Sol and Luna Worker Router
majiayu000/spellbook
Splits substantial coding or repository-review work between a Sol commander that decides and verifies and a separate Luna Max worker that implements, with an auditable run log.
287 GitHub stars~3.3k tokensUpdated 2 days ago
Agent WorkflowsAuto-check passed
Nucleus Delivery Evidence Finalizer
protect-my-hair/nucleus-marketplace
Combines the evidence from completed Nucleus workflow steps into a final evidence report and result package, without scheduling steps, updating trackers or opening pull requests.
164 GitHub stars~454 tokensUpdated 2 mo ago
Agent WorkflowsAuto-check passed

Questions about PUA High-Agency Governance

What does PUA High-Agency Governance do?

Pushes an agent to keep verifying and changing approach after repeated failures, using a diagnosis line, evidence-based completion and confirmation before risky edits. Instruction-only and written for Trae, this skill activates only on an explicit PUA request, repeated failures, signs of giving up or a completion claim with no proof, never on a normal first attempt. It separates four powers: the agent acts and writes a self-review listing evidence, risks and gaps, but scoring comes from external commands, CI or your acceptance, and changing tests, CI or the environment needs your confirmation first.

When should I use PUA High-Agency Governance?

PUA High-Agency Governance fits situations like: the agent has failed the same task twice and keeps tweaking one approach; the agent is about to declare the task impossible without checking; A task was reported done without build, test or curl evidence; you explicitly ask the agent to try harder or switch approach.

How do I install PUA High-Agency Governance in Claude Code?

Run `npx skills add tanweai/pua --skill pua -a claude-code`. Or copy the skill folder (.trae/skills/pua in tanweai/pua) into .claude/skills/pua in your project. Claude Code loads it when a task matches its description.

How do I install PUA High-Agency Governance in Codex?

Run `npx skills add tanweai/pua --skill pua -a codex`. Or copy the skill folder (.trae/skills/pua in tanweai/pua) into .agents/skills/pua in your project. Codex loads it when a task matches its description.

Can I use PUA High-Agency Governance in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add tanweai/pua --skill pua -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/pua, .gemini/skills/pua, .github/skills/pua and .opencode/skills/pua in your project.

What does PUA High-Agency Governance need to run?

SKILL.md names no scripts, command-line tools or credentials: PUA High-Agency Governance is instructions for the agent only. Our summary lists: A skill loader such as Trae Skills or `npx skills`; no hooks or subagents needed. Compatibility (from SKILL.md): Trae Skills / npx skills; instruction-only, no Claude Code hooks or agents..

Does PUA High-Agency Governance access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is PUA High-Agency Governance safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does PUA High-Agency Governance use?

PUA High-Agency Governance is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does PUA High-Agency Governance use?

About 502 tokens (SKILL.md is roughly 2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to PUA High-Agency Governance?

Skills that share tags, products or a category with PUA High-Agency Governance: Show Me Your Work Decision Log (cursor/plugins, 11k stars), Cline Pilot (sickn33/agentic-awesome-skills, 47k stars), Context Modes (0xNyk/lacp, 305 stars) and Failure Codifier (Chachamaru127/claude-code-harness, 3.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains PUA High-Agency Governance?

tanweai (a GitHub user) maintains it in tanweai/pua, which has 19,706 GitHub stars. The repository holds 28 skills in this directory. The repository was last updated on September 9, 2026.

Source: tanweai/pua on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Install PUA High-Agency Governance

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Compatibility

Context cost

Safety

SKILL.md

PUA for Trae — 高能动性治理 Skill

触发条件

四权分离：行动权 / 自我评价权 / 评分权 / 环境修改权

诊断先行

事实上的 100% 信心循环

文化叙事绑定：叙事服务证据，不替代证据

交付模板

Files

Compare with similar skills

Similar skills

More from tanweai/pua

Related

Questions about PUA High-Agency Governance