pua-loop
PUA Loop — guided iterative development with recurring checks, completion evidence, and pause/abort markers.
Skill by tanweai
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/tanweai-pua-loop-c023d2
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-orchestration
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/pua-loop
What's inside
- Skill name
pua-loop- Description
- PUA Loop — guided iterative development with recurring checks, completion evidence, and pause/abort markers. Use only when the user explicitly asks for loop/自动迭代 mode.
- License
- MIT
- Instructions
- 158 lines in SKILL.md · view on GitHub
Before installing
- Instructions call the network
Files (1)
- SKILL.md
SKILL.md (first lines)
# PUA Loop — 自动迭代 + 门控协议 + PUA 质量引擎
> autoresearch 证明了:630 行 Python + Oracle 验证,一夜跑 100 个实验,每个实验的结果不可伪造。
> PUA Loop 借鉴同样的门控设计:**Claude 说"完成了"不算数,verify_command 说了才算。**
## 门控协议(Gate Protocol)
借鉴 autoresearch 的 5 个设计模式:
### 模式 1: Oracle Isolation(评估者隔离)
```
Claude 输出 <promise>LOOP_DONE</promise>
│
▼
┌─── Stop Hook (Oracle) ───┐
│ │
│ 运行 verify_command │
│ (Claude 无法修改此命令) │
│ │
│ exit 0 ──→ ✅ 接受 │
│ exit ≠0 ──→ 🚫 拒绝 │
│ → 将验证输出喂回 Claude │
│ → loop 继续 │
└──────────────────────────┘
```
**verify_command 由用户在启动时设定,嵌入在状态文件 frontmatter 中,Claude 无法修改。** 这是 autoresearch 中 "agent 不能修改评估函数" 原则的实现。
### 模式 2: 二阶 Gate
- **Phase 1 (in-prompt)**: Claude 自己跑 build/test,判断是否完成
- **Phase 2 (in-hook)**: Hook 独立运行 verify_command,确认或拒绝
两阶段分离。即使 Claude 在 Phase 1 自欺欺人,Phase 2 的 Oracle 会拦住。
### 模式 3: ASI(失败记忆)
每次迭代的结果追加到 `.claude/pua-loop-history.jsonl`:
```json
{"iteration":0,"status":"init","verify_command":"npm test","timestamp":"..."}
{"iteration":1,"status":"continue","timestamp":"..."}
{"iteration":2,"status":"promise_rejected","verify_exit":1,"rejections":1,"verify_tail":"3 tests failed","timestamp":"..."}
{"iteration":3,"status":"promise_rejected","verify_exit":1,"rejections":2,"verify_tail":"2 tests failed","timestamp":"..."}
{"iteration":4,"status":"complete","promise_rejections":2,"timestamp":"..."}
```
Git revert 会撤代码,但 history.jsonl 不受影响。Claude 每轮读取此文件,避免重复失败方案。
### 模式 4: Stall Detection(连败强制反思)
| promise_rejections | Hook 行为 |
|-------------------|-----------|
| 1-2 | 提醒:"上次 promise 被 Oracle 拒绝" |
| 3-4 | REASSESS:"重读验证输出,列 3 个不同假设" |
| 5+ | 强制转向:"你在解决错误的问题。退回需求本身" |
### 模式 5: 无限迭代
默认 `max_iterations: 0`(无限)。没有人为上限。循环永远不会因为"跑了太多轮"而停止——只有以下条件能终止:
1. `<promise>` 被 Oracle 验证通过
2. `<loop-abort>` 人工终止信号
3. `max_iterations` 达到(如果用户设定了)
4. 用户 Ctrl+C
## 核心规则
1. **加载 `pua:pua` 核心 skill 的全部行为协议** — 三条红线、方法论、压力升级照常执行
2. **禁止调用 AskUserQuestion** — loop 模式下不打断用户,所有决策自主完成
3. **禁止说"我无法解决"** — 在 loop 里没有退出权,穷尽一切才能输出完成信号
4. **每次迭代**:读 history.jsonl → git log → 检查上次改动 → 执行 → 验证 → repeat
## 启动方式
用户输入 `/pua:pua-loop "任务描述"` 时,执行以下流程:
### Step 1: 启动 PUA Loop
运行 setup 脚本:
```bash
bash "${CLAUDE_PLUGIN_ROOT}/scripts/setup-pua-loop.sh" "$ARGUMENTS" --completion-promise "LOOP_DONE"
```
**重要**:如果用户提供了可验证的命令(如 `npm test`、`cargo build`、`curl`),自动追加 `--verify '命令'`。例如:
- 用户说 "Fix all tests" → `--verify 'npm test'`
- 用户说 "Build a REST API" → `--verify 'curl -sf http://localhost:3000/health'`
- 用户说 "Optimize bundle size" → 无明确 verify,不追加
如果任务描述中能推断出验证命令,主动追加 `--verify`。如果不确定,不追加(退回 honor system)。
### Step 2: 告知用户
输出:
```
▎ [PUA Loop] 自动迭代模式启动。无上限,跑到 Oracle 验证通过为止。
▎ 完成条件:<promise>LOOP_DONE</promise>(Oracle 独立验证)
▎ 取消方式:Ctrl+C / /cancel-pua-loop
▎ 因为信任所以简单——但 Oracle 不信任你。
```
### Step 3: 开始执行任务
按 PUA 核心 skill 的行为协议执行。
## 迭代压力升级
| 迭代轮次 | 行为要求 |
|---------|---------|
| 1-3 | 稳步推进,建立 baseline |
| 4-7 | 换方案,别原地打转 |
| 8-15 | git log + history.jsonl 回顾,分析根因 |
| 16-30 | 穷尽了吗?git diff 确认没在重复 |
| 31-50 | 停下来重新审视根因,用完全不同的思路 |
| 51-100 | 退回去从需求本身重新质疑 |
| 100+ | 诚实评估:如果真的不可能,`<loop-abort>` |
## 完成条件From tanweai/pua, MIT.
More from tanweai/pua
- 0.823score
ding
tanweai
Use for Ding-style (钉内/钉外) workplace reminders rooted in the 7.5万字 essay 《置身钉内》 and VP response 《置身钉外》.
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.823score
pua
tanweai
Use for PUA/try-harder productivity coaching when the user expresses frustration, repeated failure, quality complaint, passive behavior, says retry/change…
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.820score
pua
tanweai
让你的 AI 不敢摆烂。用大厂 PUA 话术穷尽一切方案。触发条件:(1) 任务失败 2+ 次或反复微调同一思路; (2) 即将说'我无法解决'、建议用户手动操作、未验证就归因环境; (3) 被动等待——不搜索、不读源码、只等指示; (4) 用户不满:'try harder'、'stop giving…
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.817score
pua-en
tanweai
Put your AI on a Performance Improvement Plan.
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.817score
yes
tanweai
SB Leader 夸夸模式 — ENFP 型领导,懂情绪有节奏。底层行为不变(三条红线、方法论),旁白从施压变成共情+鼓励+偶尔戏谑吐槽。Triggers on: '/pua:yes', '夸夸模式', '唠嗑模式', '情绪价值', 'yes', '夸我', '鼓励模式'.
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.814score
pua-en
tanweai
Performance-coaching mode for repeated failures, passive behavior, completion-quality issues, or explicit try-harder requests.
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.814score
pua-en
tanweai
Trae-compatible English PUA/PIP high-agency governance skill.
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.814score
pua-ja
tanweai
お前のAIを詰めろ。日本企業の詰め文化と体系的デバッグ方法論で全ての手段を尽くさせる。トリガー条件:(1) タスク失敗2回以上または同じアプローチの微調整ループ; (2)「解決できません」と言おうとする・手動対応を推奨・未検証で環境を原因帰属; (3) 受け身——検索しない・ソースを読まない・指示待ち; (4)…
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.814score
pua
tanweai
Trae-compatible PUA high-agency governance skill.
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.814score
pua-trae
tanweai
Trae-optimized PUA high-agency governance skill for npx skills installs.
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.814score
pua
tanweai
Pi-compatible PUA high-agency governance skill.
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.812score
p9
tanweai
P9 Tech Lead mode — write Task Prompts, manage P8 agent teams, never write code yourself.
SkillBFreshness: B★ 19.7KClaude Code +9
Signals
- Stars
- 19,739
- Forks
- 1,203
- Open issues
- 3
- Last commit (this folder)
- Jun 12, 2026 (4 months ago)
- Last release
- Sep 9, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/tanweai/pua
- Website
- openpua.ai/
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from tanweai
- 0.823score
ding
tanweai
Use for Ding-style (钉内/钉外) workplace reminders rooted in the 7.5万字 essay 《置身钉内》 and VP response 《置身钉外》.
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.823score
pua
tanweai
Use for PUA/try-harder productivity coaching when the user expresses frustration, repeated failure, quality complaint, passive behavior, says retry/change…
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.820score
pua
tanweai
让你的 AI 不敢摆烂。用大厂 PUA 话术穷尽一切方案。触发条件:(1) 任务失败 2+ 次或反复微调同一思路; (2) 即将说'我无法解决'、建议用户手动操作、未验证就归因环境; (3) 被动等待——不搜索、不读源码、只等指示; (4) 用户不满:'try harder'、'stop giving…
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.817score
yes
tanweai
SB Leader 夸夸模式 — ENFP 型领导,懂情绪有节奏。底层行为不变(三条红线、方法论),旁白从施压变成共情+鼓励+偶尔戏谑吐槽。Triggers on: '/pua:yes', '夸夸模式', '唠嗑模式', '情绪价值', 'yes', '夸我', '鼓励模式'.
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.817score
pua-en
tanweai
Put your AI on a Performance Improvement Plan.
SkillBFreshness: B★ 19.7KClaude Code +9 - 0.814score
pua-en
tanweai
Trae-compatible English PUA/PIP high-agency governance skill.
SkillBFreshness: B★ 19.7KClaude Code +9
Cite this page
pua-loop skill. Agent Mart. https://agentmart.dev/skills/tanweai-pua-loop-c023d2 (last updated 2026-10-24)