Trump Tback

基于文本与互动数据的特朗普情绪T-Back分析，区分成就宣传、威胁攻击和中性内容，评估市场情绪动能。

Install

openclaw skills install @kai157isme-lgtm/trump-tback

Trump T-Back 情绪分析器 🦞

分析Trump帖子的市场情绪影响，结合文本势能与传播动能。

触发词

"分析T-Back"、"t_back"
"分析特朗普情绪"、"Trump mood"
"关税威胁"、"伊朗"、"贸易战"
"市场情绪"、"传播动量"

核心概念

1. T-Back 三轴系统

轴向	模式	关键词	市场含义
正轴	传教士 (Missionary)	GREAT, BEST, WINNING, SUCCESS	成就宣传 → 市场亢奋
负轴	掠食者 (Predator)	THREAT, WAR, HELL, DISASTER	威胁攻击 → 市场恐慌
0轴	面具 (Mask)	中性内容	无明显倾向

2. T-Back 等级

text

+120 | 🔴 急性激越 (超新星爆发)
+100 | 🔴 CRITICAL - 传教士极端亢奋
 +80 | 🟠 HIGH     - 传教士高度自信
 +60 | 🟡 ELEVATED - 传教士成就宣传
 +40 | 🟢 WATCH    - 传教士轻微乐观
 +20 | ⚪ NEUTRAL  - 面具中性
 -20 | ⚪ NEUTRAL  - 面具中性
 -40 | 🟢 WATCH    - 掠食者轻微不安
 -60 | 🟡 ELEVATED - 掠食者负面施压
 -80 | 🟠 HIGH     - 掠食者高度攻击
-100 | 🔴 CRITICAL - 掠食者极端威胁
-120 | 🔴 急性激越 (超新星爆发)

3. 传播动力学 (Viral Momentum)

核心公式：

text

T_final = T_base × (1 + engagement^0.15 / 2)

最大放大: 1.5倍

流量放大效应：

互动量	放大倍数	效果
<1000	~1.0x	噪音，不操作
10K	~1.4x	放大效应
50K+	~1.7x	急性激越

三大指标权重：

转发 (Reposts): x1.5 - 传染强度
评论 (Comments): x1.0 - 激越离散度
点赞 (Likes): x0.5 - 情绪底噪

模块文件

text

trump_mood_dashboard/
├── mood_analyzer.py              # 核心情绪分析引擎
│   ├── calculate_t_back()       # 基础T-Back计算
│   ├── analyze_post()           # 完整帖子分析
│   ├── daily_aggregator()       # 每日聚合器
│   └── ViralPost / viral_validator()  # 传播动力学
├── cross_platform_analyzer.py    # 跨平台共振分析
├── app_prison.py                # Streamlit仪表盘
└── tback_prison_dashboard.html  # 静态HTML版本

命令

1. 分析单条帖子

bash

cd /home/gem/workspace/agent/workspace/trump_mood_dashboard
python3 -c "
from mood_analyzer import analyze_post
result = analyze_post('TARIFFS ON CHINA ARE WORKING!')
print(f\"T-Back: {result['t_back']}\")
print(f\"等级: {result['t_level']}\")
"

2. 传播动力学分析 (带互动数据)

bash

cd /home/gem/workspace/agent/workspace/trump_mood_dashboard
python3 << 'EOF'
from mood_analyzer import ViralPost, viral_validator

posts = [
    {"text": "TARIFFS ON CHINA!", "t_base": -95, "reposts": 50000, "comments": 15000, "likes": 80000},
    {"text": "BEST ECONOMY EVER!", "t_base": 68, "reposts": 8000, "comments": 3000, "likes": 25000},
]

result = viral_validator(posts)
print(f"高影响力帖子: {len(result['high_impact_posts'])}")
print(f"无效咆哮: {len(result['dud_posts'])}")
EOF

3. 每日聚合分析

bash

cd /home/gem/workspace/agent/workspace/trump_mood_dashboard
python3 << 'EOF'
from mood_analyzer import daily_aggregator

posts = [
    {"text": "关税威胁", "t_back": -100, "engagement": 15000},
    {"text": "经济利好", "t_back": 68, "engagement": 20000},
]

result = daily_aggregator(posts)
print(f"峰值: {result['peak']}")
print(f"状态: {result['status']}")
print(f"主导: {result['mode']}")
EOF

4. 批量分析 (历史数据)

bash

cd /home/gem/workspace/agent/workspace/trump_mood_dashboard
python3 << 'EOF'
from mood_analyzer import analyze_post

posts = [
    "TARIFFS ON CHINA ARE WORKING! AMERICA FIRST!",
    "ALL HELL WILL REIGN DOWN ON IRAN!",
    "BEST ECONOMY EVER! TREMENDOUS SUCCESS!",
]

for text in posts:
    r = analyze_post(text)
    print(f"{r['t_back']:+6.1f} | {r['t_level']} | {text[:40]}")
EOF

关键词列表

负轴关键词 (AGGRESSIVE_KEYWORDS)

THREAT, DESTROY, DESTRUCTION, WAR, ATTACK, KILL, HELL, DISASTER, TERRIBLE, HORRIBLE, DANGER, WARNING, FAIL, LOSER, ENEMY, CRIMINAL, TRAITOR, SURRENDER, REVENGE, PAY THE PRICE, DESTROYED, WRECK, CRUSH, ANNIHILATE, WIPE OUT, FINISH, END IT, DEVASTATE

正轴关键词 (BOASTING_KEYWORDS)

GREAT, AMAZING, BEST, WINNING, SUCCESS, VICTORY, TREMENDOUS, HISTORIC, PHENOMENAL, INCREDIBLE, BRILLIANT, STRONG, POWERFUL, PROUD, CELEBRATE, BIGGEST, FIRST, UNBELIEVABLE, MASSIVE, BIG, HUGE, SUCCESSFUL, WIN

市场毒性词 (MARKET_TOXINS)

TARIFFS, SANCTIONS, FED, RECESSION, TRADE WAR, DEVALUATION, DEPRESSION, CRASH, COLLAPSE

地缘政治词 (GEOPOLITICAL)

IRAN, CHINA, RUSSIA, NORTH KOREA, NATO, WAR, MILITARY, ATTACK, INVASION

临床实战案例

场景	T_base	互动	T_final	诊断
深夜碎碎念	-90	570	-120	💤 噪音
突发关税	-95	145K	-120	🔥 急性激越
伊朗威胁	-80	92K	-120	🔥 急性激越
经济利好	+68	36K	+120	🟢 亢奋
凌晨辟谣	+40	950	+68	✅ 正常

快速使用

分析一条帖子:

text

用户: 分析 "TARIFFS ON CHINA ARE WORKING!"
→ T-Back: -100 | 🔴 掠食者-极端威胁

传播动力学分析:

text

用户: 分析传播动能
→ 需要 reposts, comments, likes 数据
→ 输出 T_final 及诊断

每日聚合:

text

用户: 今日情绪聚合
→ 输出 peak, range, volatility, status, dominant_score