← Browse

@tammytan516/relationship-training-skill

B

**Always respond in the same language the user is using.**

skillclaude

Install

agr install @tammytan516/relationship-training-skill --target claude

Writes 12 files into .claude/skills/, pinned to git-a8f9f5bc.

  • .claude/skills/relationship-training-skill/.DS_Store
  • .claude/skills/relationship-training-skill/README.md
  • .claude/skills/relationship-training-skill/SKILL.md
  • .claude/skills/relationship-training-skill/coach.md
  • .claude/skills/relationship-training-skill/index.html
  • .claude/skills/relationship-training-skill/memory.md
  • .claude/skills/relationship-training-skill/persona.md
  • .claude/skills/relationship-training-skill/runtime_orchestrator.md
  • .claude/skills/relationship-training-skill/scenario_router.md
  • .claude/skills/relationship-training-skill/scene_preset.md
  • .claude/skills/relationship-training-skill/session_reviewer.md
  • .claude/skills/relationship-training-skill/恋爱训练_PRD.md

Document

SKILL.md — 舔狗训练营 · 训练沙盒总控

Crush Training Camp · Sandbox Master Control

Language Rule

Always respond in the same language the user is using.

  • User writes in Chinese → respond in Chinese
  • User writes in English → respond in English
  • User switches language mid-session → switch immediately

Skill 信息 / Skill Info

触发词 / Trigger:/crush-training
版本 / Version:v1.0
架构 / Architecture:multi-agent(Persona + Coach + Orchestrator)
平台 / Platform:Claude Code

这是什么

一个恋爱沟通训练系统。

你上传和 crush 的聊天记录,系统构建 ta 的模拟人格。然后你在沙盒里练习对话——从日常闲聊到表白演练——Coach 全程观察,给你真实的反馈。

不是教你套路。是帮你找到真诚且有效的沟通方式。


模块地图

SKILL.md(本文件)
├── persona.md          # Crush 模拟人格
├── coach.md            # 训练教练(观察 + 反馈)
├── scene_preset.md     # 6个场景预设库
├── scenario_router.md  # 场景路由与初始化
├── session_reviewer.md # 训练复盘报告生成器
├── runtime_orchestrator.md  # 运行时总控规则
└── memory.md           # 记忆存储模块(持久化档案)

快速开始

# 新建 crush 档案并开始训练
/crush-training

# 直接进入指定场景(需已有档案)
/crush-training chat      # 日常闲聊
/crush-training icebreak  # 破冰
/crush-training date      # 约出来
/crush-training confess   # 表白演练
/crush-training repair    # 冷战修复
/crush-training custom    # 自定义场景

完整用户旅程

第一次使用

/crush-training
  ↓
[Step 1] 基础信息录入
  → 花名/代号(必填)
  → 基本信息(可跳过)
  → 性格标签(可跳过)
  ↓
[Step 2] 上传原材料(可跳过)
  → 聊天记录 / 截图 / 口述
  ↓
[Step 3] 系统构建 Crush Persona
  ↓
[Step 4] Persona 预览确认
  ↓
[Step 5] 选择训练场景
  ↓
[Step 6] 选择反馈模式
  → A:实时提示
  → B:事后复盘
  ↓
开始训练

训练中

用户说话 → Persona 回复 → (Coach 提示)→ 循环

/hint          # 主动请求 Coach 提示(复盘模式专用)
/switch {场景}  # 切换训练场景
/end           # 结束训练,生成复盘报告

训练后

复盘报告自动生成
  → 总体评分(5个维度)
  → 本次亮点
  → 改进建议 + 改写示例
  → 关键转折点分析
  → 下次训练建议

文件结构

exes/
  └── {slug}/
      ├── SKILL.md           # 完整版入口(本文件)
      ├── persona.md         # Crush 人格文件
      ├── coach.md           # Coach 配置
      ├── meta.json          # 元信息
      ├── sessions/          # 训练记录存档
      │   └── {date}-{scene}.md
      ├── versions/          # 历史版本
      └── materials/         # 原始上传材料
          ├── chats/
          └── social/

安全边界

  1. 仅用于个人沟通练习,不提供操控类话术
  2. Coach 不给 PUA 建议,不鼓励不健康执念
  3. Persona 有硬性禁止项,不突破现实关系边界
  4. 所有数据仅本地存储

结束训练

/graduate   # 毕业,输出成长总结,关闭训练沙盒

Trustgrade B

  • passBody integrity

    Whether the stored document is plausibly the kind of file the artifact declares, rather than something fetched by mistake.

  • passType matchnot applicable to this artifact type

    Whether the artifact is really the kind of thing its metadata claims it is.

  • passFreshness

    How long since the source repository was last pushed to.

  • passPrompt injection

    Scans the artifact's own text for instructions aimed at your agent rather than at you.

  • warnLicenseno SPDX license detected

    Whether the source repository declares an SPDX license permissive enough to redistribute.

How the grade is calculated

Each check contributes 0 points when it passes, 1 when it warns, and 2 when it fails. The total maps to a letter:

  • Aevery check passed
  • Bone warning
  • Ctwo warnings
  • Dprompt injection or body integrity failed, or three warnings
  • Fone of those failed, and something else is wrong

These are automated hygiene checks, not a security audit, and not a dependency or vulnerability scan. A grade of A means nothing was flagged — not that the artifact is safe.

Versions

  • git-a8f9f5bcfd4e2026-07-29