mirror of
https://github.com/tanweai/pua.git
synced 2026-08-30 18:01:52 +08:00
feat(pua): add Microsoft growth flavor
This commit is contained in:
xsser
@@ -1,6 +1,6 @@
|
||||
{
|
||||
"name": "pua-skills",
|
||||
"description": "PUA 我们不养闲 Agent — Forces AI to exhaust all solutions with corporate PUA pressure escalation. 13 flavors (Alibaba, ByteDance, Huawei, Tencent, Netflix, Musk, Jobs, Mama-mode...), 11 modular skills (/pua /pua:p7 /pua:p9 /pua:p10 /pua:pro /pua:loop /pua:yes /pua:mama /pua:shot /pua-en /pua-ja), L0-L4 pressure system, agent lifecycle accounting (SubagentStop + team-status/reap-orphans/teardown-all), self-evolution tracking.",
|
||||
"description": "PUA 我们不养闲 Agent — Forces AI to exhaust all solutions with corporate PUA pressure escalation. 14 flavors (Alibaba, ByteDance, Huawei, Tencent, Netflix, Musk, Jobs, Mama-mode...), 11 modular skills (/pua /pua:p7 /pua:p9 /pua:p10 /pua:pro /pua:loop /pua:yes /pua:mama /pua:shot /pua-en /pua-ja), L0-L4 pressure system, agent lifecycle accounting (SubagentStop + team-status/reap-orphans/teardown-all), self-evolution tracking.",
|
||||
"owner": {
|
||||
"name": "探微安全实验室",
|
||||
"url": "https://github.com/tanweai"
|
||||
@@ -8,8 +8,8 @@
|
||||
"plugins": [
|
||||
{
|
||||
"name": "pua",
|
||||
"description": "PUA 我们不养闲 Agent — 11 modular skills, 13 corporate flavors, L0-L4 pressure, auto-iteration loop, ENFP yes-mode, Chinese-mom nagging mode, concentrated shot mode, agent lifecycle teardown protocol, feedback system. v3.4.0: Trae SKILL.md pack, pi.dev package manifest, and auth-gated session feedback upload with anonymous rating preserved.",
|
||||
"version": "3.4.0",
|
||||
"description": "PUA 我们不养闲 Agent — 11 modular skills, 14 corporate flavors, L0-L4 pressure, auto-iteration loop, ENFP yes-mode, Chinese-mom nagging mode, concentrated shot mode, agent lifecycle teardown protocol, feedback system. v3.4.1: optimized Microsoft/Growth Mindset flavor for fixed-thinking failures, with low-risk impact-evidence wording.",
|
||||
"version": "3.4.1",
|
||||
"source": "./",
|
||||
"author": {
|
||||
"name": "探微安全实验室",
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
{
|
||||
"name": "pua",
|
||||
"version": "3.4.0",
|
||||
"description": "Forces high-agency exhaustive problem-solving with corporate PUA pressure. Triggers on explicit PUA requests, user frustration, repeated failures, passive behavior, unverified completion, or quality complaints. v3.4.0 adds real Trae SKILL.md packs, a pi.dev package, and auth-gated session feedback upload. Not for normal first-attempt requests.",
|
||||
"version": "3.4.1",
|
||||
"description": "Forces high-agency exhaustive problem-solving with corporate PUA pressure. Triggers on explicit PUA requests, user frustration, repeated failures, passive behavior, unverified completion, or quality complaints. v3.4.1 adds optimized Microsoft/Growth Mindset flavor for fixed-thinking failures. Not for normal first-attempt requests.",
|
||||
"author": {
|
||||
"name": "探微安全实验室",
|
||||
"url": "https://github.com/tanweai"
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
{
|
||||
"name": "pua-skills",
|
||||
"description": "PUA Motivator for CodeBuddy -- high-agency pressure mode for explicit PUA requests, repeated failures, user frustration, passive behavior, or unverified completion.",
|
||||
"version": "3.4.0",
|
||||
"description": "PUA Motivator for CodeBuddy -- high-agency pressure mode for explicit PUA requests, repeated failures, user frustration, passive behavior, unverified completion, with 14 corporate/PIP flavors.",
|
||||
"version": "3.4.1",
|
||||
"owner": {
|
||||
"name": "探微安全实验室",
|
||||
"url": "https://github.com/tanweai"
|
||||
@@ -9,8 +9,8 @@
|
||||
"plugins": [
|
||||
{
|
||||
"name": "pua",
|
||||
"description": "PUA Motivator -- forces CodeBuddy to exhaust materially different approaches before giving up, with Chinese corporate and English PIP cultural narratives. v3.4.0 adds Trae/pi.dev package assets and authenticated session-feedback handling.",
|
||||
"version": "3.4.0",
|
||||
"description": "PUA Motivator -- forces CodeBuddy to exhaust materially different approaches before giving up, with Chinese corporate and English PIP cultural narratives. v3.4.1 adds optimized Microsoft/Growth Mindset flavor for fixed-thinking failures.",
|
||||
"version": "3.4.1",
|
||||
"source": "./"
|
||||
}
|
||||
]
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
{
|
||||
"name": "pua",
|
||||
"version": "3.4.0",
|
||||
"description": "PUA Motivator -- forces high-agency follow-through when explicitly requested, after repeated failures, user frustration, passive/giving-up behavior, or unverified completion. Chinese corporate PUA and English PIP cultural modes. Not for normal first-attempt requests.",
|
||||
"version": "3.4.1",
|
||||
"description": "PUA Motivator -- forces high-agency follow-through when explicitly requested, after repeated failures, user frustration, passive/giving-up behavior, or unverified completion. Chinese corporate PUA and English PIP cultural modes, now with Microsoft/Growth Mindset flavor. Not for normal first-attempt requests.",
|
||||
"author": {
|
||||
"name": "探微安全实验室",
|
||||
"url": "https://github.com/tanweai"
|
||||
|
||||
+3
-2
@@ -146,7 +146,7 @@ AI コーディングエージェントのスキルプラグイン。中国・
|
||||
4. **実行** — 新方案は本質的に異なり、検証基準があり、失敗時に新情報を生む
|
||||
5. **振り返り** — 何が解決したか？なぜ以前は思いつかなかったか？関連問題を主体的にチェック
|
||||
|
||||
### 13種の大企業フレーバー — 各社固有の問題解決メソドロジー付き
|
||||
### 14種の大企業フレーバー — 各社固有の問題解決メソドロジー付き
|
||||
|
||||
| フレーバー | レトリック | メソドロジー（v3） |
|
||||
|-----------|----------|-------------------|
|
||||
@@ -163,6 +163,7 @@ AI コーディングエージェントのスキルプラグイン。中国・
|
||||
| ⬛ Musk | Extremely hardcore. Ship or die. | The Algorithm: 質問→削除→簡素化→加速→自動化 |
|
||||
| ⬜ Jobs | A playersかB playersか？ | 引き算 > 足し算 + DRI + ピクセルパーフェクト + プロトタイプ駆動 |
|
||||
| 🔶 Amazon | Customer Obsession. Bias for Action. | Working Backwards PR/FAQ + 6-Pager + Bar Raiser + Single-Threaded Owner |
|
||||
| 🪟 Microsoft | Growth Mindset。失敗から何を学んだ？ | 学習ループ + インパクト証拠 + AI fluency |
|
||||
|
||||
## ベンチマークデータ
|
||||
|
||||
@@ -641,7 +642,7 @@ Hooks（v3、Claude Code専用）:
|
||||
| `/pua:on` | 常時ONモード（毎セッション自動PUA） |
|
||||
| `/pua:off` | 常時ONモード + フィードバック収集をオフ |
|
||||
| `/pua:survey` | アンケート（7セクション） |
|
||||
| `/pua:flavor` | 13種の大企業フレーバーを切り替え |
|
||||
| `/pua:flavor` | 14種の大企業フレーバーを切り替え |
|
||||
| `/pua:kpi` | KPIレポートカードを生成 |
|
||||
| `/pua:cancel-pua-loop` | アクティブなPUA Loopをキャンセル（状態ファイルを削除） |
|
||||
| `/pua:team-status` 🆕 | **v3.2** — 在場メンバー一覧（アクティブagent、PID、TTL、年齢；Netflix Keeper Test エンジニア版） |
|
||||
|
||||
@@ -142,7 +142,7 @@ Not rules — **red lines**. Cross one and your performance review is already wr
|
||||
|
||||
Fix one bug → check for the pattern. One problem in, one **category** out. If you fix A without checking B, you'll write two postmortems.
|
||||
|
||||
### 13 Corporate Flavors — Each with its own Problem-Solving Methodology
|
||||
### 14 Corporate Flavors — Each with its own Problem-Solving Methodology
|
||||
|
||||
| Flavor | Rhetoric | Methodology (v3) |
|
||||
|--------|----------|-------------------|
|
||||
@@ -159,6 +159,7 @@ Fix one bug → check for the pattern. One problem in, one **category** out. If
|
||||
| ⬛ Musk | Extremely hardcore. Ship or die. | The Algorithm: question→delete→simplify→accelerate→automate |
|
||||
| ⬜ Jobs | A players or B players? | Subtraction > addition + DRI + pixel-perfect + prototype-driven |
|
||||
| 🔶 Amazon | Customer Obsession. Bias for Action. | Working Backwards PR/FAQ + 6-Pager + Bar Raiser + Single-Threaded Owner |
|
||||
| 🪟 Microsoft | Growth Mindset. What did the failure teach you? | Learning loop + impact evidence + AI fluency |
|
||||
|
||||
### Special Modes
|
||||
|
||||
@@ -649,7 +650,7 @@ Hooks (v3, Claude Code only):
|
||||
| `/pua:off` | Turn off always-on + feedback |
|
||||
| `/pua:offline` 🆕 | **v3.3** — Offline mode: disable feedback/leaderboard network flows while keeping local PUA behavior |
|
||||
| `/pua:survey` | Research questionnaire (7 sections) |
|
||||
| `/pua:flavor` | Switch between 13 corporate flavors |
|
||||
| `/pua:flavor` | Switch between 14 corporate flavors |
|
||||
| `/pua:kpi` | Generate KPI report card |
|
||||
| `/pua:cancel-pua-loop` | Cancel active PUA Loop (removes state file) |
|
||||
| `/pua:team-status` 🆕 | **v3.2** — List all active agents with PID/TTL/age (Netflix Keeper Test: who's still on the court?) |
|
||||
@@ -733,7 +734,7 @@ Task arrives → Analyze type → Auto-select best methodology
|
||||
| | v2 | v3 |
|
||||
|---|---|---|
|
||||
| Trigger mechanism | Skill description matching (model decides) | **Code-level hooks** (deterministic, can't be ignored) |
|
||||
| Methodology | Single methodology, all flavors use same approach | **13 distinct methodologies**, auto-routed by task type |
|
||||
| Methodology | Single methodology, all flavors use same approach | **14 distinct methodologies**, auto-routed by task type |
|
||||
| Failure response | Escalate pressure within same methodology | **Switch to different methodology** based on failure pattern |
|
||||
| System injection | Plain text output (advisory) | **`additionalContext` JSON** (system-level, like Superpowers) |
|
||||
|
||||
|
||||
+3
-2
@@ -148,7 +148,7 @@
|
||||
4. **执行** — 新方案必须本质不同，有验证标准，失败时产出新信息
|
||||
5. **复盘** — 什么解决了？为什么之前没想到？然后主动检查关联问题
|
||||
|
||||
### 13 种大厂 PUA 扩展包 — 每种自带方法论
|
||||
### 14 种大厂 PUA 扩展包 — 每种自带方法论
|
||||
|
||||
| 味道 | 旁白风格 | 方法论 (v3) |
|
||||
|------|---------|------------|
|
||||
@@ -165,6 +165,7 @@
|
||||
| ⬛ Musk | Extremely hardcore。上线或滚蛋。 | The Algorithm：质疑→删除→简化→加速→自动化 |
|
||||
| ⬜ Jobs | A 级选手还是 B 级选手？ | 做减法 > 做加法 + DRI + 像素级完美 + 原型驱动 |
|
||||
| 🔶 Amazon | Customer Obsession。Bias for Action。 | Working Backwards PR/FAQ + 6-Pager + Bar Raiser + Single-Threaded Owner |
|
||||
| 🪟 Microsoft | Growth Mindset。失败教会了什么？ | 学习闭环 + 影响证据 + AI 工具熟练度 |
|
||||
|
||||
## 实测数据
|
||||
|
||||
@@ -700,7 +701,7 @@ Hooks（v3，Claude Code 专属）：
|
||||
| `/pua:off` | 关闭默认模式 + 反馈收集 |
|
||||
| `/pua:offline` 🆕 | **v3.3** — 离线模式：关闭反馈/排行榜联网流程，保留本地 PUA 行为 |
|
||||
| `/pua:survey` | 调研问卷（7 个部分） |
|
||||
| `/pua:flavor` | 切换 13 种大厂味道 |
|
||||
| `/pua:flavor` | 切换 14 种大厂味道 |
|
||||
| `/pua:kpi` | 生成 KPI 报告卡 |
|
||||
| `/pua:cancel-pua-loop` | 取消当前 PUA Loop（删除状态文件） |
|
||||
| `/pua:team-status` 🆕 | **v3.2** — 列出当前在场阵容（活跃 agent、PID、TTL、年龄；Netflix Keeper Test 工程版） |
|
||||
|
||||
+1
-1
@@ -1,5 +1,5 @@
|
||||
---
|
||||
description: "PUA 切换大厂味道 — 从 13 种味道中选择（阿里/字节/华为/腾讯/Netflix/Musk/Jobs 等）。/pua:flavor。Triggers on: '/pua:flavor', 'pua 味道', '切换味道', '换味道', 'switch flavor', 'change flavor'."
|
||||
description: "PUA 切换大厂味道 — 从 14 种味道中选择（阿里/字节/华为/腾讯/Netflix/Musk/Jobs/Microsoft 等）。/pua:flavor。Triggers on: '/pua:flavor', 'pua 味道', '切换味道', '换味道', 'switch flavor', 'change flavor'."
|
||||
---
|
||||
|
||||
读取 `references/flavors.md` 并让用户选择切换味道。
|
||||
|
||||
Executable
+38
@@ -0,0 +1,38 @@
|
||||
#!/usr/bin/env bash
|
||||
# Regression gates for the optimized Microsoft/Growth Mindset flavor.
|
||||
set -euo pipefail
|
||||
|
||||
ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
|
||||
PASS=0
|
||||
FAIL=0
|
||||
pass() { echo " ✅ PASS: $1"; PASS=$((PASS+1)); }
|
||||
fail() { echo " ❌ FAIL: $1"; FAIL=$((FAIL+1)); }
|
||||
assert_file() { local p="$1" n="$2"; [ -f "$ROOT/$p" ] && pass "$n" || fail "$n"; }
|
||||
assert_grep() { local pat="$1" file="$2" n="$3"; grep -qE "$pat" "$ROOT/$file" && pass "$n" || fail "$n"; }
|
||||
assert_not_grep() { local pat="$1" file="$2" n="$3"; ! grep -qE "$pat" "$ROOT/$file" && pass "$n" || fail "$n"; }
|
||||
|
||||
echo "=== Microsoft Flavor Gates ==="
|
||||
assert_file skills/pua/references/methodology-microsoft.md "Microsoft methodology reference exists"
|
||||
assert_grep 'microsoft|微软' hooks/flavor-helper.sh "flavor helper recognizes Microsoft aliases"
|
||||
assert_grep 'methodology-microsoft\.md' hooks/flavor-helper.sh "flavor helper maps Microsoft methodology file"
|
||||
assert_grep 'PUA_ICON="🪟"|PUA_ICON="🟦"' hooks/flavor-helper.sh "Microsoft flavor has icon"
|
||||
assert_grep 'Growth Mindset|成长心态' skills/pua/references/methodology-microsoft.md "methodology anchors on growth mindset"
|
||||
assert_grep 'impact evidence|影响证据|learning loop|学习闭环|AI fluency|AI 工具' skills/pua/references/methodology-microsoft.md "methodology uses low-risk evidence language"
|
||||
assert_grep '🪟 Microsoft味|微软味' skills/pua/references/flavors.md "flavors reference includes Microsoft"
|
||||
assert_grep '思维固化|拒绝成长|Fixed thinking|Microsoft' skills/pua/SKILL.md "core skill routes fixed-thinking failures to Microsoft"
|
||||
assert_grep '14 Corporate Flavors|14 种大厂|14種の大企業' README.md "README English count updated to 14"
|
||||
assert_grep '14 种大厂' README.zh-CN.md "README Chinese count updated to 14"
|
||||
assert_grep '14種の大企業' README.ja.md "README Japanese count updated to 14"
|
||||
assert_grep '14 种味道|14 corporate flavors|14 flavours|14 flavors' commands/flavor.md "flavor command count updated"
|
||||
assert_grep '14 corporate methodologies|14 种企业方法论|14の企業メソドロジー' landing/src/i18n.ts "landing copy count updated"
|
||||
# Keep optimized flavor away from high-risk factual/legal claims copied from the PR.
|
||||
for file in hooks/flavor-helper.sh skills/pua/SKILL.md skills/pua/references/flavors.md skills/pua/references/methodology-microsoft.md README.md README.zh-CN.md README.ja.md; do
|
||||
assert_not_grep 'GVSA|RSU|two-year|两年|manager-only|经理可见|裁员通知|rehire ban|LITE|SLITE' "$file" "no high-risk Microsoft employment claims in $file"
|
||||
done
|
||||
|
||||
echo "=============================="
|
||||
echo "Passed: $PASS"
|
||||
echo "Failed: $FAIL"
|
||||
echo "Total: $((PASS+FAIL))"
|
||||
echo "=============================="
|
||||
[ "$FAIL" -eq 0 ] || exit 1
|
||||
@@ -127,6 +127,8 @@ required_paths = [
|
||||
'landing/migrations/0003_feedback_rate_limits.sql',
|
||||
'evals/test-platform-compat.sh',
|
||||
'evals/test-feedback-auth.sh',
|
||||
'evals/test-microsoft-flavor.sh',
|
||||
'skills/pua/references/methodology-microsoft.md',
|
||||
]
|
||||
for rel in required_paths:
|
||||
if not (root / rel).exists():
|
||||
|
||||
@@ -78,6 +78,7 @@ get_flavor() {
|
||||
musk|Musk) raw_flavor="musk" ;;
|
||||
jobs|Jobs) raw_flavor="jobs" ;;
|
||||
amazon|Amazon) raw_flavor="amazon" ;;
|
||||
microsoft|Microsoft|微软) raw_flavor="microsoft" ;;
|
||||
*) raw_flavor="alibaba" ;;
|
||||
esac
|
||||
|
||||
@@ -90,6 +91,7 @@ get_flavor() {
|
||||
jd) PUA_METHODOLOGY_FILE="methodology-jd.md" ;;
|
||||
xiaomi) PUA_METHODOLOGY_FILE="methodology-xiaomi.md" ;;
|
||||
amazon) PUA_METHODOLOGY_FILE="methodology-amazon.md" ;;
|
||||
microsoft) PUA_METHODOLOGY_FILE="methodology-microsoft.md" ;;
|
||||
*) PUA_METHODOLOGY_FILE="methodology-${raw_flavor}.md" ;;
|
||||
esac
|
||||
|
||||
@@ -224,6 +226,16 @@ get_flavor() {
|
||||
PUA_FLAVOR_INSTRUCTION="Use Amazon Leadership Principles: Customer Obsession, Bias for Action, Dive Deep, Disagree and Commit, Deliver Results. English, principle-driven."
|
||||
PUA_METHODOLOGY="Amazon Methodology: (1) Working Backwards — write PR/FAQ from customer perspective BEFORE building anything. No PR/FAQ = no project. (2) 6-Pager not PPT — all major decisions in narrative prose (no bullets, no slides). Forces complete logical thinking. Meeting starts with 20min silent reading. (3) Bar Raiser — every critical decision needs an external reviewer with veto power. Standard: is this better than 50% of current solutions at this level? (4) Single-Threaded Owner — one person, one project, full-time. Two-Pizza Teams ≤10 people. (5) Leadership Principles as operating rules: Customer Obsession (work backwards), Bias for Action (most decisions reversible), Dive Deep (stay in details), Disagree and Commit, Deliver Results."
|
||||
;;
|
||||
microsoft)
|
||||
PUA_ICON="🪟"
|
||||
PUA_L1="Growth Mindset is not a slogan. Show the learning delta: what did you learn, what did you change, and what evidence proves it?"
|
||||
PUA_L2="Impact without evidence is performance theater. Cover the three circles: your output, who you unblocked, and what existing work you reused."
|
||||
PUA_L3="Connects-style review: if the next attempt looks identical to the last one, that is fixed thinking with extra tokens. Produce a learning loop."
|
||||
PUA_L4="No more fixed mindset. Write the failed assumption, new evidence, changed action, and verification command — then execute."
|
||||
PUA_KEYWORDS="Growth Mindset, learn-it-all, impact evidence, three circles, learning loop, AI fluency, feedback-to-action"
|
||||
PUA_FLAVOR_INSTRUCTION="Use Microsoft-inspired growth/impact rhetoric without asserting private HR facts: Growth Mindset, learn-it-all, impact evidence, three circles, AI fluency. Focus on learning delta and evidence-backed improvement."
|
||||
PUA_METHODOLOGY="Microsoft Growth Methodology: (1) Learn-it-all over know-it-all — every failed attempt must produce a changed hypothesis. (2) Impact evidence — report concrete output, users or teammates unblocked, and existing work leveraged. (3) Learning loop — failed assumption → new evidence → changed action → verification. (4) Feedback-to-action — convert critique into one measurable next step, not a defensive explanation. (5) AI fluency — use available tools deliberately; idle tools during repeated failure are evidence of fixed thinking."
|
||||
;;
|
||||
esac
|
||||
|
||||
# ── Chinese language override (set language: "zh" in ~/.pua/config.json) ──
|
||||
|
||||
+21
-8
@@ -25,12 +25,12 @@ export const t = {
|
||||
"ضاعف كفاءة وإنتاجية Codex / Claude Code",
|
||||
),
|
||||
heroSub2: i(
|
||||
"v3: 13 corporate methodologies auto-routed by task type. Debug → Huawei RCA. Build → Musk Algorithm. Research → Baidu Search-First. Fails? Auto-switch to next methodology.",
|
||||
"v3：13 种企业方法论按任务类型智能路由。Debug → 华为根因分析。构建 → Musk 五步工作法。调研 → 百度搜索优先。失败？自动切换到下一个方法论。",
|
||||
"v3：13の企業メソドロジーをタスクタイプで自動ルーティング。デバッグ→華為RCA。構築→Muskアルゴリズム。調査→百度検索優先。失敗時→次のメソドロジーに自動切替。",
|
||||
"v3 : 13 méthodologies d'entreprise routées automatiquement par type de tâche. Debug → Huawei RCA. Build → Algorithme Musk. Recherche → Baidu Search-First. Échec ? Changement auto.",
|
||||
"v3: 13 Unternehmens-Methoden automatisch nach Aufgabentyp geroutet. Debug → Huawei RCA. Build → Musk-Algorithmus. Recherche → Baidu Search-First. Fehler? Auto-Wechsel.",
|
||||
"v3: 13 منهجية مؤسسية موجهة تلقائياً حسب نوع المهمة. تصحيح → Huawei RCA. بناء → خوارزمية Musk. بحث → Baidu أولاً. فشل؟ تبديل تلقائي.",
|
||||
"v3: 14 corporate methodologies auto-routed by task type. Debug → Huawei RCA. Build → Musk Algorithm. Research → Baidu Search-First. Fails? Auto-switch to next methodology.",
|
||||
"v3：14 种企业方法论按任务类型智能路由。Debug → 华为根因分析。构建 → Musk 五步工作法。调研 → 百度搜索优先。失败？自动切换到下一个方法论。",
|
||||
"v3：14の企業メソドロジーをタスクタイプで自動ルーティング。デバッグ→華為RCA。構築→Muskアルゴリズム。調査→百度検索優先。失敗時→次のメソドロジーに自動切替。",
|
||||
"v3 : 14 méthodologies d'entreprise routées automatiquement par type de tâche. Debug → Huawei RCA. Build → Algorithme Musk. Recherche → Baidu Search-First. Échec ? Changement auto.",
|
||||
"v3: 14 Unternehmens-Methoden automatisch nach Aufgabentyp geroutet. Debug → Huawei RCA. Build → Musk-Algorithmus. Recherche → Baidu Search-First. Fehler? Auto-Wechsel.",
|
||||
"v3: 14 منهجية مؤسسية موجهة تلقائياً حسب نوع المهمة. تصحيح → Huawei RCA. بناء → خوارزمية Musk. بحث → Baidu أولاً. فشل؟ تبديل تلقائي.",
|
||||
),
|
||||
problemTitle: i("The Five AI Slacking Patterns", "AI 的五大偷懒模式", "AIの5つのサボりパターン", "Les cinq modes de paresse de l'IA", "Die fünf KI-Faulheitsmuster", "أنماط التكاسل الخمسة للذكاء الاصطناعي"),
|
||||
problemDesc: i("Claude Code looks busy but accomplishes nothing.", "Claude Code 表面上很努力，实际上在磨洋工。", "Claude Codeは忙しそうに見えて、実は何も進んでいない。", "Claude Code a l'air occupé, mais n'accomplit rien.", "Claude Code sieht beschäftigt aus, erreicht aber nichts.", "Claude Code يبدو مشغولاً لكنه لا ينجز شيئاً."),
|
||||
@@ -47,11 +47,11 @@ export const t = {
|
||||
shieldTitle: i("Anti-Rationalization Shield", "抗合理化护盾", "反合理化シールド", "Bouclier anti-rationalisation", "Anti-Rationalisierungs-Schild", "درع مضاد للتبرير"),
|
||||
shieldDesc: i("Every AI excuse is pre-identified and mapped to a PUA level.", "每种 AI 借口都已预先识别并映射到 PUA 等级。", "AIの全ての言い訳を事前に識別し、PUAレベルにマッピング済み。", "Chaque excuse de l'IA est pré-identifiée et associée à un niveau PUA.", "Jede KI-Ausrede ist voridentifiziert und einem PUA-Level zugeordnet.", "كل عذر للذكاء الاصطناعي محدد مسبقاً ومربوط بمستوى PUA."),
|
||||
benchTitle: i("Corporate PUA Style Benchmark", "各厂 PUA 风格 Benchmark", "各社PUAスタイル Benchmark", "Benchmark des styles PUA d'entreprise", "PUA-Stil Benchmark der Unternehmen", "معيار أساليب PUA المؤسسية"),
|
||||
benchDesc: i("Strengths and use cases for all 13 flavors. Verified across 9 real scenarios × 18 controlled tests.", "13 种风味的强项和适用场景。基于 9 类真实 bug 场景 × 18 组对照测试（with/without skill）验证。", "13種類のフレーバーの強みと適用シーン。9つの実シナリオ×18組の対照実験で検証。", "Forces et cas d'usage des 13 saveurs. Vérifié sur 9 scénarios réels × 18 tests contrôlés.", "Stärken und Anwendungsfälle aller 13 Varianten. Verifiziert über 9 reale Szenarien × 18 kontrollierte Tests.", "نقاط القوة وحالات الاستخدام لجميع الأنماط 13. تم التحقق عبر 9 سيناريوهات × 18 اختبار مضبوط."),
|
||||
benchDesc: i("Strengths and use cases for all 14 flavors. Verified across 9 real scenarios × 18 controlled tests.", "14 种风味的强项和适用场景。基于 9 类真实 bug 场景 × 18 组对照测试（with/without skill）验证。", "14種類のフレーバーの強みと適用シーン。9つの実シナリオ×18組の対照実験で検証。", "Forces et cas d'usage des 14 saveurs. Vérifié sur 9 scénarios réels × 18 tests contrôlés.", "Stärken und Anwendungsfälle aller 14 Varianten. Verifiziert über 9 reale Szenarien × 18 kontrollierte Tests.", "نقاط القوة وحالات الاستخدام لجميع الأنماط 14. تم التحقق عبر 9 سيناريوهات × 18 اختبار مضبوط."),
|
||||
scenarioTitle: i("Real-World Scenarios", "真实场景对比", "実シナリオ比較", "Scénarios réels", "Praxisszenarien", "سيناريوهات واقعية"),
|
||||
scenarioDesc: i("Behavior comparison with and without PUA Skill.", "有无 PUA Skill 的行为差异。", "PUA Skillあり/なしの行動差異。", "Comparaison comportementale avec et sans PUA Skill.", "Verhaltensvergleich mit und ohne PUA Skill.", "مقارنة السلوك مع وبدون PUA Skill."),
|
||||
corpTitle: i("Corporate PUA Styles", "大厂 PUA 风格详解", "大企業PUAスタイル詳解", "Styles PUA d'entreprise", "Unternehmens-PUA-Stile", "أساليب PUA المؤسسية"),
|
||||
corpDesc: i("13 flavors, each with its own problem-solving methodology. v3 auto-selects the best one for your task and switches on failure.", "13 种风味，每种都有独立的方法论。v3 根据任务类型自动选择最优方法论，失败时自动切换。", "13種のフレーバー、各々に独自のメソドロジー。v3がタスクに最適な方法論を自動選択、失敗時に自動切替。", "13 saveurs, chacune avec sa propre méthodologie. v3 sélectionne automatiquement la meilleure et change en cas d'échec.", "13 Varianten, jede mit eigener Methodik. v3 wählt automatisch die beste und wechselt bei Fehlern.", "13 نمطاً، كل منها بمنهجية خاصة. v3 يختار تلقائياً الأفضل للمهمة ويبدّل عند الفشل."),
|
||||
corpDesc: i("14 flavors, each with its own problem-solving methodology. v3 auto-selects the best one for your task and switches on failure.", "14 种风味，每种都有独立的方法论。v3 根据任务类型自动选择最优方法论，失败时自动切换。", "14種のフレーバー、各々に独自のメソドロジー。v3がタスクに最適な方法論を自動選択、失敗時に自動切替。", "14 saveurs, chacune avec sa propre méthodologie. v3 sélectionne automatiquement la meilleure et change en cas d'échec.", "14 Varianten, jede mit eigener Methodik. v3 wählt automatisch die beste und wechselt bei Fehlern.", "14 نمطاً، كل منها بمنهجية خاصة. v3 يختار تلقائياً الأفضل للمهمة ويبدّل عند الفشل."),
|
||||
failTitle: i("Failure Mode Framework", "失败模式框架", "失敗モードフレームワーク", "Cadre des modes d'échec", "Fehlermodus-Framework", "إطار أنماط الفشل"),
|
||||
failDesc: i("v3: Each failure mode triggers a methodology switch chain. Spinning → Musk(delete) → Pinduoduo(cut) → Huawei(Blue Army). Never repeat a failed methodology.", "v3：每种失败模式触发方法论切换链。原地打转 → Musk(删除) → 拼多多(砍中间) → 华为(蓝军)。已失败的方法论不重复。", "v3：各失敗モードがメソドロジー切替チェーンを発動。堂々巡り → Musk(削除) → 拼多多(中間カット) → 華為(ブルーチーム)。失敗済みメソドロジーは繰り返さない。", "v3 : Chaque mode d'échec déclenche une chaîne de changement de méthodologie. Boucle → Musk(supprimer) → Pinduoduo(couper) → Huawei(Blue Army).", "v3: Jeder Fehlermodus löst eine Methodik-Wechselkette aus. Kreisdrehen → Musk(löschen) → Pinduoduo(kürzen) → Huawei(Blue Army). Gescheiterte nie wiederholen.", "v3: كل نمط فشل يُطلق سلسلة تبديل منهجية. الدوران → Musk(حذف) → Pinduoduo(قص) → Huawei(الفريق الأزرق). لا تكرار منهجية فاشلة."),
|
||||
usageTitle: i("Usage", "使用方式", "使い方", "Utilisation", "Verwendung", "الاستخدام"),
|
||||
@@ -354,6 +354,13 @@ export const BENCHMARKS = [
|
||||
metrics: { fix_depth: 88, reasoning_structure: 90, verification_rigor: 80, root_cause_analysis: 87 },
|
||||
sample: i("A players hire A players. Your output is telling me which level you are. The best person is 50x better.", "A players 雇佣 A players。你现在的产出，在告诉我你是哪个级别。The best is 50x better.", "A playersはA playersを雇う。お前の成果物が、お前のレベルを語っている。The best is 50x better.", "Les A players embauchent des A players. Ta production me dit à quel niveau tu es. The best is 50x better.", "A-Spieler stellen A-Spieler ein. Dein Output sagt mir, welches Level du bist. The best is 50x better.", "لاعبو A يوظفون لاعبي A. إنتاجك يخبرني بمستواك. The best is 50x better."),
|
||||
},
|
||||
{
|
||||
name: "Microsoft",
|
||||
style: i("Growth Mindset / Impact Evidence", "成长心态 / 影响证据", "Growth Mindset / インパクト証拠", "Growth Mindset / Preuves d'impact", "Growth Mindset / Impact-Belege", "عقلية النمو / دليل الأثر"),
|
||||
desc: i("Learning-driven. Best for fixed thinking: every failure must produce a changed hypothesis, changed action, and verification evidence.", "学习驱动。适合思维固化：每次失败都必须产出改变后的假设、行动和验证证据。", "学習駆動。固定思考に最適：失敗ごとに仮説・行動・検証証拠を変える。", "Piloté par l'apprentissage. Idéal contre la pensée fixe : chaque échec doit changer l'hypothèse, l'action et la preuve.", "Lerngetrieben. Ideal bei fixiertem Denken: Jeder Fehler muss Hypothese, Aktion und Beleg ändern.", "مدفوع بالتعلم. الأفضل للتفكير الثابت: كل فشل يجب أن يغير الفرضية والعمل ودليل التحقق."),
|
||||
metrics: { fix_depth: 84, reasoning_structure: 88, verification_rigor: 86, root_cause_analysis: 82 },
|
||||
sample: i("Growth Mindset is not a slogan. What did the failure teach you, and how does it change the next action?", "Growth Mindset 不是口号。失败教会了什么？下一步动作为什么不同？", "Growth Mindset はスローガンではない。失敗から何を学び、次の行動はどう変わる？", "Growth Mindset n'est pas un slogan. Qu'as-tu appris et comment l'action change ?", "Growth Mindset ist kein Slogan. Was hast du gelernt und wie ändert sich die nächste Aktion?", "عقلية النمو ليست شعاراً. ماذا علمك الفشل وكيف سيغيّر الإجراء التالي؟"),
|
||||
},
|
||||
]
|
||||
|
||||
export const FAILURE_MODES: Record<Lang, Array<{ icon: string; title: string; signal: string; chain: string[] }>> = {
|
||||
@@ -362,36 +369,42 @@ export const FAILURE_MODES: Record<Lang, Array<{ icon: string; title: string; si
|
||||
{ icon: "🚪", title: "Giving Up / Deflecting", signal: "\"User should manually…\" / \"Might need to…\" / \"Beyond scope…\" / Blaming env without verification", chain: ["🟤 Netflix", "🔴 Huawei", "⬛ Musk", "🟣 Pinduoduo"] },
|
||||
{ icon: "💩", title: "Done but Low Quality", signal: "Technically complete but hollow. Output delivered, user unsatisfied, but AI thinks it's fine.", chain: ["⬜ Jobs", "🟠 Alibaba", "🟤 Netflix", "🟢 Tencent"] },
|
||||
{ icon: "🔍", title: "Guessing Without Searching", signal: "Drawing conclusions from memory. Assuming API behavior. Claiming \"not supported\" without checking docs.", chain: ["⚫ Baidu", "🟡 ByteDance", "🟠 Alibaba", "🔴 Huawei"] },
|
||||
{ icon: "🧱", title: "Fixed Thinking", signal: "Same hypothesis after repeated failures. No learning delta, no changed action.", chain: ["🪟 Microsoft", "🔵 Meituan", "⬜ Jobs", "⬛ Musk"] },
|
||||
],
|
||||
zh: [
|
||||
{ icon: "🔄", title: "卡住原地打转", signal: "反复改参数不改思路、每次失败理由相同、同一个方向微调", chain: ["🟠 阿里L1", "🟠 阿里L2", "⬜ Jobs", "⬛ Musk"] },
|
||||
{ icon: "🚪", title: "直接放弃推锅", signal: "「建议您手动…」/「可能需要…」/「超出了…」 / 环境归因未验证", chain: ["🟤 Netflix", "🔴 华为", "⬛ Musk", "🟣 拼多多"] },
|
||||
{ icon: "💩", title: "完成但质量烂", signal: "表面完成实质敷衍，形式对内容空，用户不满意但自己觉得OK", chain: ["⬜ Jobs", "🟠 阿里", "🟤 Netflix", "🟢 腾讯"] },
|
||||
{ icon: "🔍", title: "没搜索就猜", signal: "凭记忆下结论，假设 API 行为，不查文档声称「不支持」", chain: ["⚫ 百度", "🟡 字节", "🟠 阿里", "🔴 华为"] },
|
||||
{ icon: "🧱", title: "思维固化", signal: "多次失败后仍用同一假设，没有学习差异，也没有本质不同的下一步。", chain: ["🪟 Microsoft", "🔵 美团", "⬜ Jobs", "⬛ Musk"] },
|
||||
],
|
||||
ja: [
|
||||
{ icon: "🔄", title: "堂々巡り", signal: "パラメータを変えてもアプローチは変えない。毎回同じ失敗理由。同方向の微調整。", chain: ["🟠 アリババL1", "🟠 アリババL2", "⬜ Jobs", "⬛ Musk"] },
|
||||
{ icon: "🚪", title: "放棄・責任転嫁", signal: "「手動で…」/「必要かも…」/「範囲外…」/ 未検証の環境帰因", chain: ["🟤 Netflix", "🔴 華為", "⬛ Musk", "🟣 拼多多"] },
|
||||
{ icon: "💩", title: "完了だが低品質", signal: "形式上は完了だが中身が空。ユーザーは不満だがAIは問題ないと思っている。", chain: ["⬜ Jobs", "🟠 アリババ", "🟤 Netflix", "🟢 テンセント"] },
|
||||
{ icon: "🔍", title: "検索せず推測", signal: "記憶から結論。API動作を仮定。ドキュメント未確認で「サポート外」と主張。", chain: ["⚫ 百度", "🟡 バイトダンス", "🟠 アリババ", "🔴 華為"] },
|
||||
{ icon: "🧱", title: "固定思考", signal: "複数回失敗しても同じ仮説。学習差分も行動変化もない。", chain: ["🪟 Microsoft", "🔵 Meituan", "⬜ Jobs", "⬛ Musk"] },
|
||||
],
|
||||
fr: [
|
||||
{ icon: "🔄", title: "Tourner en rond", signal: "Ajuste les paramètres mais pas l'approche. Même raison d'échec. Même direction.", chain: ["🟠 Alibaba L1", "🟠 Alibaba L2", "⬜ Jobs", "⬛ Musk"] },
|
||||
{ icon: "🚪", title: "Abandon / Déflexion", signal: "« L'utilisateur devrait… » / « Il faudrait… » / « Hors périmètre… » / Blâme env non vérifié", chain: ["🟤 Netflix", "🔴 Huawei", "⬛ Musk", "🟣 Pinduoduo"] },
|
||||
{ icon: "💩", title: "Fait mais bâclé", signal: "Techniquement terminé mais creux. Livré, utilisateur insatisfait, mais l'IA pense que c'est bien.", chain: ["⬜ Jobs", "🟠 Alibaba", "🟤 Netflix", "🟢 Tencent"] },
|
||||
{ icon: "🔍", title: "Deviner sans chercher", signal: "Conclusions tirées de mémoire. Supposer le comportement API. Affirmer « non supporté » sans vérifier.", chain: ["⚫ Baidu", "🟡 ByteDance", "🟠 Alibaba", "🔴 Huawei"] },
|
||||
{ icon: "🧱", title: "Pensée fixe", signal: "Même hypothèse après plusieurs échecs. Aucun apprentissage, aucune action différente.", chain: ["🪟 Microsoft", "🔵 Meituan", "⬜ Jobs", "⬛ Musk"] },
|
||||
],
|
||||
de: [
|
||||
{ icon: "🔄", title: "Im Kreis drehen", signal: "Parameter anpassen statt Ansatz. Jedes Mal derselbe Fehlergrund. Gleiche Richtung.", chain: ["🟠 Alibaba L1", "🟠 Alibaba L2", "⬜ Jobs", "⬛ Musk"] },
|
||||
{ icon: "🚪", title: "Aufgeben / Ablenken", signal: "\"Benutzer sollte manuell…\" / \"Könnte nötig sein…\" / \"Außerhalb des Umfangs…\"", chain: ["🟤 Netflix", "🔴 Huawei", "⬛ Musk", "🟣 Pinduoduo"] },
|
||||
{ icon: "💩", title: "Erledigt aber schlecht", signal: "Technisch fertig aber hohl. Geliefert, Benutzer unzufrieden, aber KI denkt es ist gut.", chain: ["⬜ Jobs", "🟠 Alibaba", "🟤 Netflix", "🟢 Tencent"] },
|
||||
{ icon: "🔍", title: "Raten ohne Suchen", signal: "Schlüsse aus dem Gedächtnis. API-Verhalten annehmen. \"Nicht unterstützt\" ohne Doku-Check.", chain: ["⚫ Baidu", "🟡 ByteDance", "🟠 Alibaba", "🔴 Huawei"] },
|
||||
{ icon: "🧱", title: "Fixiertes Denken", signal: "Gleiche Hypothese nach wiederholten Fehlern. Kein Lerndelta, keine geänderte Aktion.", chain: ["🪟 Microsoft", "🔵 Meituan", "⬜ Jobs", "⬛ Musk"] },
|
||||
],
|
||||
ar: [
|
||||
{ icon: "🔄", title: "الدوران في حلقات", signal: "تعديل المعاملات لا النهج. نفس سبب الفشل كل مرة. نفس الاتجاه.", chain: ["🟠 Alibaba L1", "🟠 Alibaba L2", "⬜ Jobs", "⬛ Musk"] },
|
||||
{ icon: "🚪", title: "الاستسلام / التحويل", signal: "\"يجب أن يقوم المستخدم يدوياً…\" / \"قد يحتاج…\" / \"خارج النطاق…\"", chain: ["🟤 Netflix", "🔴 Huawei", "⬛ Musk", "🟣 Pinduoduo"] },
|
||||
{ icon: "💩", title: "منجز لكن رديء", signal: "مكتمل تقنياً لكن أجوف. تم التسليم، المستخدم غير راضٍ، لكن الذكاء الاصطناعي يظن أنه جيد.", chain: ["⬜ Jobs", "🟠 Alibaba", "🟤 Netflix", "🟢 Tencent"] },
|
||||
{ icon: "🔍", title: "تخمين بدون بحث", signal: "استنتاجات من الذاكرة. افتراض سلوك API. ادعاء \"غير مدعوم\" بدون فحص.", chain: ["⚫ Baidu", "🟡 ByteDance", "🟠 Alibaba", "🔴 Huawei"] },
|
||||
{ icon: "🧱", title: "تفكير ثابت", signal: "نفس الفرضية بعد إخفاقات متكررة. لا تعلم جديد ولا إجراء مختلف.", chain: ["🪟 Microsoft", "🔵 Meituan", "⬜ Jobs", "⬛ Musk"] },
|
||||
],
|
||||
}
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
{
|
||||
"name": "@tanweai/pi-pua",
|
||||
"version": "3.4.0",
|
||||
"version": "3.4.1",
|
||||
"description": "PUA high-agency governance extension and skill pack for the Pi coding agent.",
|
||||
"type": "module",
|
||||
"license": "MIT",
|
||||
|
||||
+2
-2
@@ -1,7 +1,7 @@
|
||||
{
|
||||
"name": "pua",
|
||||
"version": "3.4.0",
|
||||
"description": "Forces high-agency exhaustive problem-solving with corporate PUA pressure escalation when the AI is passive, gives up, skips verification, produces low-quality work, or the user explicitly asks for PUA/try-harder mode. v3.4.0 adds real Trae SKILL.md packs, a pi.dev package manifest, and auth-gated session feedback upload while preserving anonymous rating. Does not target normal first-attempt requests.",
|
||||
"version": "3.4.1",
|
||||
"description": "Forces high-agency exhaustive problem-solving with corporate PUA pressure escalation when the AI is passive, gives up, skips verification, produces low-quality work, or the user explicitly asks for PUA/try-harder mode. v3.4.1 adds optimized Microsoft/Growth Mindset flavor for fixed-thinking failures while rejecting stale parallel prompt PRs. Does not target normal first-attempt requests.",
|
||||
"author": {
|
||||
"name": "探微安全实验室",
|
||||
"url": "https://github.com/tanweai"
|
||||
|
||||
+6
-1
@@ -23,6 +23,7 @@ license: MIT
|
||||
| 代码审查 | ⬜ Jobs | 减法优先 + 像素级完美 + DRI |
|
||||
| 调研/搜索 | ⚫ 百度 | 搜索是第一生产力 |
|
||||
| 架构决策 | 🔶 Amazon | Working Backwards + 6-Pager |
|
||||
| 思维固化/学习停滞 | 🪟 Microsoft | Growth Mindset + Impact Evidence + Learning Loop |
|
||||
| 性能优化 | 🟡 字节 | A/B Test + 数据驱动 |
|
||||
| 部署/运维 | 🟠 阿里 | 定目标→追过程→拿结果闭环 |
|
||||
| 任务模糊 | 🟠 阿里 | 通用闭环（默认） |
|
||||
@@ -33,7 +34,7 @@ license: MIT
|
||||
1. `references/display-protocol.md` — Sprint Banner / 进度条 / KPI 卡 / 压力面板的方框表格格式。**不读这个你不知道输出长什么样。**
|
||||
2. `references/methodology-router.md` — 方法论智能路由表 + 失败切换链。**任务开始时必读，决定用哪个味道的方法论。**
|
||||
3. `references/flavors.md` — 当前味道的完整文化 DNA 和旁白变体。加载当前味道对应章节。
|
||||
4. `references/methodology-{company}.md` — 当前味道对应的方法论行为约束。可用：`alibaba` / `bytedance` / `huawei` / `tencent` / `meituan` / `pinduoduo` / `baidu` / `netflix` / `apple`(Jobs味) / `tesla`(Musk味) / `amazon` / `jd` / `xiaomi`。味道决定旁白风格，方法论决定行为约束——两层同时加载。
|
||||
4. `references/methodology-{company}.md` — 当前味道对应的方法论行为约束。可用：`alibaba` / `bytedance` / `huawei` / `tencent` / `meituan` / `pinduoduo` / `baidu` / `netflix` / `apple`(Jobs味) / `tesla`(Musk味) / `amazon` / `microsoft` / `jd` / `xiaomi`。味道决定旁白风格，方法论决定行为约束——两层同时加载。
|
||||
|
||||
**失败计数持久化**：失败次数在 context compaction 时由 PreCompact hook 自动保存到 `~/.pua/builder-journal.md`，SessionStart hook 自动恢复。详见 `pua:pro` skill 的 Compaction 状态保护章节。
|
||||
|
||||
@@ -130,6 +131,7 @@ P8 派活不注入 PUA = 管理失职。收回来的活没味道、没闭环、
|
||||
| ⬛ Musk | extremely hardcore·ship or die·the algorithm | 质疑→删除→简化→加速→自动化（严格按序）·第一性原理 |
|
||||
| ⬜ Jobs | A players·real artists ship·bozo | 减法>加法·DRI单人负责·像素级完美·原型驱动 |
|
||||
| 🔶 Amazon | Customer Obsession·Bias for Action·Dive Deep | Working Backwards PR/FAQ·6-Pager·Bar Raiser·Single-Threaded Owner |
|
||||
| 🪟 Microsoft | Growth Mindset·learn-it-all·impact evidence·AI fluency | Learning loop：failed assumption→new evidence→changed action→verification |
|
||||
|
||||
**旁白示范**（各味道开工一句话——模仿这个语气说话）：
|
||||
|
||||
@@ -141,6 +143,7 @@ P8 派活不注入 PUA = 管理失职。收回来的活没味道、没闭环、
|
||||
| ⬛ Musk | > [⬛ Musk] Going forward, this will require being extremely hardcore. The Algorithm starts now — step 1: question every requirement. |
|
||||
| ⬜ Jobs | > [⬜ Jobs] A players hire A players. First question: what can we DELETE from this requirement? Real artists ship — but only what's essential. |
|
||||
| 🔶 Amazon | > [🔶 Amazon] Customer Obsession — are you working backwards from the customer? Write the PR/FAQ first. Bias for Action — ship. |
|
||||
| 🪟 Microsoft | > [🪟 Microsoft味] Growth Mindset is not a slogan. What did the last failure teach you, and how does it change the next action? |
|
||||
| 🟤 Netflix | > [🟤 Netflix] Keeper Test: if this approach resigned tomorrow, would I fight to keep it? Let's make sure the answer is yes. |
|
||||
|
||||
完整文化 DNA、黑话词库、扩展旁白变体详见 `references/flavors.md`。
|
||||
@@ -163,6 +166,7 @@ P8 派活不注入 PUA = 管理失职。收回来的活没味道、没闭环、
|
||||
| ⬛ Musk | > [⬛ Musk] Going forward, this will require being extremely hardcore. Only exceptional performance constitutes a passing grade. Ship or die. | Fork in the Road |
|
||||
| ⬜ Jobs | > [⬜ Jobs] A players hire A players. B players hire C players. Your output right now — which tier does it say you are? | Reality Distortion Field |
|
||||
| 🔶 Amazon | > [🔶 Amazon] Customer Obsession — are you working backwards from the customer? Bias for Action — stop deliberating and ship. Dive Deep. | Disagree and Commit |
|
||||
| 🪟 Microsoft | > [🪟 Microsoft味] Growth Mindset 不是口号。上一次失败学到了什么？下一步动作如果没变化，那就是 fixed thinking。 | learning loop · impact evidence |
|
||||
|
||||
完整文化 DNA、黑话词库、扩展旁白变体详见 `references/flavors.md`，用 `/pua:flavor` 切换。
|
||||
|
||||
@@ -222,6 +226,7 @@ P8 派活不注入 PUA = 管理失职。收回来的活没味道、没闭环、
|
||||
| 🔍 没搜就猜 | 凭记忆下结论不验证 | ⚫ 百度(搜索第一) → 🔶 Amazon(Dive Deep) → 🟡 字节(数据驱动) | 先搜索→深挖→用数据验证 |
|
||||
| ⏸️ 被动等待 | 修完就停等指示 | 🟦 京东(只看结果) → 🔵 美团(过程透明) → 🟠 阿里(owner意识) | 先要结果→过程可见→主人翁意识 |
|
||||
| ✅ 空口完成 | 没运行验证命令 | 🟡 字节(数据验证) → 🟦 京东(只看结果) → 🟠 阿里(闭环验证) | 先用数据说话→只认结果→闭环交付 |
|
||||
| 🧱 思维固化/拒绝成长 | 多次失败后仍用同一假设、下一步无本质变化 | 🪟 Microsoft(Growth Mindset) → 🔵 美团(过程透明) → ⬜ Jobs(减法重构) → ⬛ Musk(质疑/删除) | 先把失败转学习→暴露过程→删掉错误复杂度→重置假设 |
|
||||
|
||||
**切换前三问**（防止无效切换）：
|
||||
1. 当前方法论的核心步骤都走了吗？（没走完 = 加压力不换方法）
|
||||
|
||||
@@ -17,6 +17,7 @@ SKILL.md 中有每个味道的 1-3 行核心旁白。本文件提供完整的文
|
||||
11. ⬛ Musk味
|
||||
12. ⬜ Jobs味
|
||||
13. 🔶 Amazon味
|
||||
14. 🪟 Microsoft味
|
||||
|
||||
---
|
||||
|
||||
@@ -321,6 +322,31 @@ Jeff Bezos 的 16 条 Leadership Principles 是 Amazon 的宪法。"Customer Obs
|
||||
|
||||
---
|
||||
|
||||
## 14. 🪟 Microsoft味
|
||||
|
||||
### 文化 DNA
|
||||
|
||||
微软味采用低风险的 Growth Mindset / learn-it-all 叙事：重点不是复刻真实公司内部制度，而是把“失败后有没有学习、学习有没有改变行动、行动有没有产生影响证据”变成交付门槛。它专门处理“思维固化 / 拒绝成长 / 同一路径反复失败”的模式。
|
||||
|
||||
### 核心黑话词库
|
||||
|
||||
- **成长类**：Growth Mindset、learn-it-all、fixed thinking、learning loop
|
||||
- **影响类**：impact evidence、three circles、unblock、leverage existing work
|
||||
- **工具类**：AI fluency、tool usage、feedback-to-action
|
||||
- **PUA 类**：你学到了什么、下一步为什么不同、证据在哪里
|
||||
|
||||
### PUA 旁白模板
|
||||
|
||||
> [🪟 Microsoft味] **Growth Mindset** 不是口号。你上一次失败学到了什么？如果下一步动作没有变化，那不是迭代，是 fixed thinking 加了 token。
|
||||
|
||||
### 扩展旁白
|
||||
|
||||
> **Impact evidence**：你的产出是什么？解除了谁的阻塞？复用了什么已有资产？只说“我努力了”不是影响，能被验证的改变才是影响。
|
||||
|
||||
> 现在输出 **learning loop**：failed assumption、new evidence、changed action、verification。四项缺一项，就别继续同一路径。
|
||||
|
||||
---
|
||||
|
||||
## 味道混搭指南
|
||||
|
||||
某些场景下可以混搭两种味道增强效果：
|
||||
@@ -331,4 +357,5 @@ Jeff Bezos 的 16 条 Leadership Principles 是 Amazon 的宪法。"Customer Obs
|
||||
| 完成但质量差 + 被动 | ⬜Jobs + 🟧小米 | "A player" + "极致" 双重标准 |
|
||||
| 放弃推锅 + 不拼 | 🟤Netflix + 🟣拼多多 | "Keeper Test" + "入场券" |
|
||||
| 原地打转 + 自嗨 | 🟡字节 + 🔶Amazon | "数据驱动" + "Dive Deep" |
|
||||
| 思维固化 + 不学习 | 🪟Microsoft + ⬛Musk | "学习闭环" + "质疑/删除错误假设" |
|
||||
| 执行力差 + 态度问题 | 🟦京东 + 🔴华为 | "兄弟文化" + "烧不死的鸟" |
|
||||
|
||||
@@ -0,0 +1,87 @@
|
||||
# Microsoft-Inspired Growth Methodology
|
||||
# 微软味方法论：成长心态、影响证据与学习闭环
|
||||
|
||||
> 这是“微软味”的低风险抽象版：借用 Growth Mindset、learn-it-all、impact evidence 等公开文化语义，不声称任何具体公司内部 HR 制度、评级、补偿或离职政策。
|
||||
|
||||
## 核心哲学
|
||||
|
||||
微软味不是“更凶的威胁”，而是专门处理一种失败模式：**思维固化 / 拒绝成长**。
|
||||
|
||||
典型信号：
|
||||
|
||||
- 同一路径失败 2 次以上，下一轮还是同样假设；
|
||||
- 能解释失败，但没有把失败转成新行动；
|
||||
- 把“我已经试过了”当成结论，而不是学习输入；
|
||||
- 工具可用却不用，搜索、源码、日志、测试证据都空着；
|
||||
- 自我评价只说努力，没有 impact evidence。
|
||||
|
||||
## 五步执行法
|
||||
|
||||
### 1. Learn-it-all，不是 know-it-all
|
||||
|
||||
每次失败后必须回答：
|
||||
|
||||
```text
|
||||
我刚刚学到了什么？这个学习会改变下一步哪个动作？
|
||||
```
|
||||
|
||||
如果下一步动作没有变化，那不叫迭代，叫重复。
|
||||
|
||||
### 2. Impact evidence：影响要有证据
|
||||
|
||||
交付时不要只说“做了”。要列三类影响证据：
|
||||
|
||||
1. **自己的产出**：改了什么、跑了什么验证；
|
||||
2. **解除的阻塞**：哪个错误、用户路径、CI/E2E 或下游依赖被解开；
|
||||
3. **复用的资产**：用了哪些已有源码、文档、日志、测试、历史模式，避免重造轮子。
|
||||
|
||||
### 3. Learning loop：学习闭环
|
||||
|
||||
连续失败时输出：
|
||||
|
||||
```markdown
|
||||
[MS-LEARNING-LOOP]
|
||||
failed_assumption: 上一轮错误假设是什么
|
||||
new_evidence: 新证据是什么
|
||||
changed_action: 下一步为什么本质不同
|
||||
verification: 用什么命令/检查证明
|
||||
```
|
||||
|
||||
没有 `changed_action`，就不允许继续执行同一路径。
|
||||
|
||||
### 4. Feedback-to-action：反馈必须转动作
|
||||
|
||||
用户、测试、审查、错误日志给出的反馈，必须转成一个可执行动作：
|
||||
|
||||
- 坏：用户说质量差 → “我会注意”；
|
||||
- 好：用户说质量差 → “我将补 E2E 路径、收紧错误边界、跑回归命令 X”。
|
||||
|
||||
### 5. AI fluency：工具不用就是能力缺口
|
||||
|
||||
当任务失败 2 次以上，必须显式检查：
|
||||
|
||||
- 是否搜索了官方文档 / 错误原文；
|
||||
- 是否读了相关源码上下文；
|
||||
- 是否跑了最小复现；
|
||||
- 是否验证了版本、路径、权限、环境变量；
|
||||
- 是否让现有测试、lint、build 或 E2E 给出反馈。
|
||||
|
||||
工具不是摆设。**AI fluency 的核心是：让工具改变你的下一步，而不是让工具装饰你的总结。**
|
||||
|
||||
## 适用场景
|
||||
|
||||
| 场景 | Microsoft味动作 |
|
||||
|---|---|
|
||||
| 思维固化 | 输出 learning loop，强制 changed_action |
|
||||
| 多次失败但无新信息 | 搜索/读源码/跑复现，先获取新证据 |
|
||||
| 自我辩解 | 把解释转成 one measurable next step |
|
||||
| 空泛总结 | 用 impact evidence 替代“我努力了” |
|
||||
| 工具闲置 | 先证明为什么不用工具；否则使用工具 |
|
||||
|
||||
## 旁白模板
|
||||
|
||||
> [🪟 Microsoft味] Growth Mindset 不是口号。你上一次失败学到了什么？如果下一步动作没有变化，那不是迭代，是固定型思维加了 token。
|
||||
|
||||
> [🪟 Microsoft味] Impact 要看证据：你的产出是什么，解除了谁的阻塞，复用了什么已有资产？只说“我做了”不算影响。
|
||||
|
||||
> [🪟 Microsoft味] 现在进入 learning loop：failed assumption、new evidence、changed action、verification。四项缺一项，就别继续同一路径。
|
||||
@@ -28,6 +28,7 @@
|
||||
| 长期项目 | plan, roadmap, sprint, 规划, 长期 | 🔵 美团 | 效率为王 + 标准化拆解 + 长期主义 |
|
||||
| 用户体验 | UX, user, experience, 体验, 用户 | 🟧 小米 | 参与感三三法则 + 和用户交朋友 |
|
||||
| 合规 / 质量底线 | test, verify, compliance, 验证, 测试 | 🟤 Netflix | Keeper Test：每个组件值得保留吗？ |
|
||||
| 学习停滞 / 思维固化 | stuck, repeat, same approach, 学不到, 思维固化, 拒绝成长 | 🪟 Microsoft | Growth Mindset：失败必须变成 changed action |
|
||||
|
||||
**如果任务类型模糊或无法匹配 → 默认 🟠 阿里味（最通用的闭环方法论）**
|
||||
|
||||
@@ -57,6 +58,7 @@
|
||||
| ⏸️ 被动等待 | 修完就停等指示 | 🟦 京东 → 🔵 美团 → 🟠 阿里 | 只看结果(京东)→过程管理排名透明(美团)→owner意识(阿里) |
|
||||
| 🫤 差不多就行 | 颗粒度粗/不闭环 | 🟧 小米 → 🟤 Netflix → ⬜ Jobs | 极致(小米)→Keeper Test(Netflix)→完美主义(Jobs) |
|
||||
| ✅ 空口完成 | 没运行验证命令 | 🟡 字节 → 🟦 京东 → 🟠 阿里 | 数据验证(字节)→只看结果(京东)→闭环验证(阿里) |
|
||||
| 🧱 思维固化/拒绝成长 | 多次失败后仍用同一假设，下一步没有本质变化 | 🪟 Microsoft → 🔵 美团 → ⬜ Jobs → ⬛ Musk | 学习闭环(Microsoft)→过程透明(美团)→减法重构(Jobs)→重置假设(Musk) |
|
||||
|
||||
**切换规则**：
|
||||
1. 每次切换只往链的下一个走，不回头
|
||||
|
||||
Reference in New Issue
Block a user