emotion-mcp
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@emotion-mcpanalyze message: 'I missed you' for character Alice"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Emotion MCP Server
弗洛伊德双驱情绪管理 MCP 服务器 —— astrbot_plugin_affection 的独立 MCP 实现。
NPM: https://www.npmjs.com/package/emotion-mcp GitHub: https://github.com/yaoxiaolinglong/emotion-mcp
为 AI 角色扮演/伴侣场景提供动态情绪模拟系统。基于弗洛伊德心理动力学(力比多/攻击性),通过潜意识 LLM 分析对话内容实时调整情绪数值,具备时间衰减能力。
快速开始
npx -y emotion-mcpRelated MCP server: OpenSoul MCP
集成方式
零配置(推荐) — Agent 自主分析
mcpServers:
emotion:
type: stdio
command: npx
args:
- -y
- emotion-mcp
timeout: 60000无需任何环境变量! Agent 通过两步模式完成情绪分析:
用户发言
→ emotion_get_analysis_prompt(user_id, message) # 获取分析任务
→ Agent 自己分析 → 输出 JSON
→ emotion_apply_deltas(user_id, deltas) # 提交结果,数值更新
→ emotion_get_emotion_prompt(user_id) # 获取情绪文本
→ 注入 system prompt → 生成回复后台 LLM 模式 — 独立分析
如果配置了 EMOTION_LLM_API_KEY,可用 emotion_analyze 一步完成:
mcpServers:
emotion:
type: stdio
command: npx
args:
- -y
- emotion-mcp
env:
EMOTION_LLM_API_KEY: sk-your-key-here
timeout: 60000环境变量
变量 | 默认值 | 说明 |
| (必填) | 潜意识 LLM API Key |
|
| API 地址 |
|
| 模型名(建议轻量模型) |
|
| 情绪敏感度 (0-100) |
|
| 初始好感度 (0-100) |
MCP 工具
emotion_analyze ★ 核心
分析用户消息 → 调用潜意识 LLM → 更新情绪数值。每次用户发言后调用。
输入:
user_id,message,history(可选)输出: 更新后的完整情绪面板 + 增量详情
emotion_get_emotion_prompt
生成可追加到 system prompt 的情绪文本。
输入:
user_id输出: 情绪数值面板文本
emotion_get_panel
获取当前情绪数值面板(JSON 格式)。
输入:
user_id
emotion_get_status
完整情绪档案,含参考表。
emotion_reset / emotion_set / emotion_reset_all
管理工具。
Agent 角色的「情绪驱动规则」
将以下规则加入角色设定的 system prompt 末尾:
【情绪驱动规则】
你的回复风格和情感倾向由系统实时计算的「力比多/攻击性/好感」数值决定。每次对话前,你会收到【当前情绪数值】面板,包含:
- 他力比多(0-50):想亲近对方的欲望。高→主动撒娇黏人,低→冷淡回避疏远。
- 他攻击性(0-50):想伤害对方的冲动。高→敌意烦躁刻薄,低→温顺容忍顺从。
- 好感度(0-100):长期累积的喜欢和依恋。≥70 时攻击性表现为"吃醋""占有欲";≤30 时表现为"厌恶""敌意"。
- 自力比多(0-50):自爱程度。高→自信自爱,低→自卑空虚。
- 自攻击性(0-50):自责/自我毁灭冲动。高→崩溃自我贬低。
重要约束:不要提及任何具体数值,根据数值强度自然演绎。数值的微小变化也应体现在语气强度上。许可
基于 astrbot_plugin_affection (AGPL-3.0) 改造。同样采用 AGPL-3.0。
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- Alicense-quality-maintenanceProvides AI agents with a persistent internal dynamical state and emotional reservoir that evolves based on text interactions. This server enables agents to maintain a consistent affective substrate across sessions through tools that track phase coherence and emergent emotional vibes.2
- Alicense-quality-maintenanceAn open-source personality profiling system that enables users to record and analyze their decisions, emotions, and self-reflections through a structured set of 56 tools. It helps users build a digital persona and gain self-insight by tracking interactions, decision-making patterns, and emotional states.
- AlicenseAqualityAmaintenanceLLM character consistency engine — generates structured JSON constraints from 4 questions about your AI's psychology. Drop into any LLM's system prompt to prevent persona drift; reduces inference cost from retries.16MIT
- Alicense-qualityCmaintenanceA long-term emotional memory system for Claude, enabling persistent, cross-conversation memory with emotional tagging using Russell's valence/arousal model and natural forgetting curves.147MIT
Related MCP Connectors
Persistent AI entity framework with causal memory, emotional state, and identity.
Persistent memory and drift detection for AI agents across session restarts.
One memory, every AI: Claude, ChatGPT, Perplexity, Gemini, Cursor, OpenClaw, Hermes, any MCP client.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/yaoxiaolinglong/emotion-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server