emotion-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| EMOTION_LLM_MODEL | No | 模型名（建议轻量模型） | gpt-3.5-turbo |
| EMOTION_LLM_API_KEY | No | 潜意识 LLM API Key（可选，用于后台LLM模式） | |
| EMOTION_SENSITIVITY | No | 情绪敏感度 (0-100) | 30 |
| EMOTION_LLM_API_BASE | No | API 地址 | https://api.mulanteastory.cn |
| EMOTION_INITIAL_AFFECTION | No | 初始好感度 (0-100) | 50 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| emotion_get_panelB | 获取用户当前情绪数值面板，返回所有情绪维度的当前值和基线值 |
| emotion_get_emotion_promptA | 生成可注入 system prompt 的情绪数值文本，让 AI 根据当前情绪数值演绎角色 |
| emotion_get_statusA | 获取用户完整情绪档案：好感度、力比多/攻击性、情感标签、对话轮次、完整参考表 |
| emotion_get_analysis_promptA | 获取情绪分析任务 — 返回当前数值 + 分析指南 + 消息和对话历史。调用后你会得到一个分析任务，请根据指南判断情绪变化并输出 JSON，然后调用 emotion_apply_deltas 提交结果。这样就由你（主 LLM）来做潜意识分析，无需额外配置 API Key |
| emotion_apply_deltasA | 提交情绪增量 JSON 以更新数值。配合 emotion_get_analysis_prompt 使用：分析完成后将 JSON 结果提交于此 |
| emotion_analyzeA | 【一键模式】分析用户消息 + 更新情绪。如果配置了 EMOTION_LLM_API_KEY 则由后台 LLM 分析，否则由调用方（Agent）分析。推荐使用 emotion_get_analysis_prompt + emotion_apply_deltas 两步模式以获得更好的控制 |
| emotion_resetA | 重置指定用户的情绪数值至初始状态 |
| emotion_setA | 手动设置用户情绪数值。未指定的字段保持不变 |
| emotion_reset_allA | ⚠️ 清除所有用户情绪档案，重置机器人自身情绪 |
| emotion_apply_decayB | 手动触发时间衰减（通常自动执行） |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 10 tools
Several get tools overlap in the data they return. get_panel and get_status both provide current emotion values, and get_analysis_prompt also includes current values, which could confuse agents. However, descriptions clearly differentiate their broader purposes (UI panel, full profile, analysis task), so overlaps are manageable.
All tools use the emotion_ prefix with snake_case, and verbs like get, apply, set, reset are consistent and predictable. The pattern is verb_noun or verb, with only minor exceptions like emotion_analyze and emotion_reset_all, but these are still clear and uniform.
10 tools is well within the typical 3-15 range and each tool serves a distinct function in managing emotion state. The number feels appropriate for a focused emotion management server, covering read, update, reset, and analysis operations without unnecessary bloat.
The server covers the full emotion lifecycle: reading (get_panel, get_status, get_analysis_prompt), updating (set, apply_deltas, analyze, apply_decay), and resetting (reset, reset_all). Minor gaps like listing all users are not essential for typical role-play scenarios, so the surface is almost complete.