Skip to main content
AllMCPs
BrowseBestCategoriesStackCompareToolsGuidesBlog
Log in Submit MCP

Stay in the loop

Get new MCP servers and top picks in your inbox.

AllMCPs

The open directory for discovering and installing Model Context Protocol servers.

AllMCPs on GitHub (opens in a new tab)
Launched onTiny Startupstinystartups.com
Explore
  • Browse servers
  • Best MCP servers
  • Categories
  • MCP clients
  • Agent prompts
  • Stack Builder
  • Compare servers
  • Random discovery New
  • Submit a server
  • Pricing & Boost Boost
Learn
  • Guides hub
  • What is MCP?
  • Install guide
  • Build an MCP server
  • Deploy an MCP server
  • Security guide
  • Troubleshooting
  • MCP for SEO & AEO
  • Protocol versioning
  • Transports: stdio vs HTTP
  • State of MCP (stats)
  • Blog & updates
Tools
  • All developer tools
  • Config generator
  • Config validator
  • Config auditor
  • MCP playground
  • Token calculator
  • OpenAPI → MCP
  • Badge generator
For agents
  • REST API docs
  • Trust & traffic Live
  • Remote MCP server SSE ↗ (opens in a new tab)
  • llms.txt ↗ (opens in a new tab)
  • Catalog JSON ↗ (opens in a new tab)
Company
  • About
  • Advertise Sponsor
  • Contact
  • GitHub ↗ (opens in a new tab)
  • Terms
  • Privacy
AllMCPs VerifiedAllMCPs VerifiedFeatured on Nick LaunchesFeatured on Nick LaunchesLaunch Llama NewsletterLaunch Llama NewsletterVerified DR - allmcps.comVerified DR - allmcps.comFeatured on SaaSGrowFeatured on SaaSGrowFeatured on Twelve ToolsFeatured on Twelve ToolsFeatured on Saaspa.geFeatured on Saaspa.geFeatured on Findly.toolsFeatured on Findly.toolsFeatured on Startup FameFeatured on Startup FameFeatured on LaunchKiwiFeatured on LaunchKiwiFeatured on ScrollLaunchFeatured on ScrollLaunchFeatured on DailyPingsFeatured on DailyPingsFazier badgeFazier badgeFeatured on NewTool.siteFeatured on NewTool.siteFeatured on saasfame.comFeatured on saasfame.comDR Checker - Domain RatingDR Checker - Domain RatingListed on Turbo0Listed on Turbo0Launched on LaunchBoard - Product Launch PlatformLaunched on LaunchBoard - Product Launch PlatformList on SimilarlabsList on Similarlabshttps://codetrendy.comhttps://codetrendy.comListed on DevTool.ioFeatured on BuildlistFeatured on BuildlistLaunched on Tiny StartupsFeatured on ShowMeBestAIFeatured on ShowMeBestAIFind us on LaunchZoneFind us on LaunchZoneAllMCPs VerifiedAllMCPs VerifiedFeatured on Nick LaunchesFeatured on Nick LaunchesLaunch Llama NewsletterLaunch Llama NewsletterVerified DR - allmcps.comVerified DR - allmcps.comFeatured on SaaSGrowFeatured on SaaSGrowFeatured on Twelve ToolsFeatured on Twelve ToolsFeatured on Saaspa.geFeatured on Saaspa.geFeatured on Findly.toolsFeatured on Findly.toolsFeatured on Startup FameFeatured on Startup FameFeatured on LaunchKiwiFeatured on LaunchKiwiFeatured on ScrollLaunchFeatured on ScrollLaunchFeatured on DailyPingsFeatured on DailyPingsFazier badgeFazier badgeFeatured on NewTool.siteFeatured on NewTool.siteFeatured on saasfame.comFeatured on saasfame.comDR Checker - Domain RatingDR Checker - Domain RatingListed on Turbo0Listed on Turbo0Launched on LaunchBoard - Product Launch PlatformLaunched on LaunchBoard - Product Launch PlatformList on SimilarlabsList on Similarlabshttps://codetrendy.comhttps://codetrendy.comListed on DevTool.ioFeatured on BuildlistFeatured on BuildlistLaunched on Tiny StartupsFeatured on ShowMeBestAIFeatured on ShowMeBestAIFind us on LaunchZoneFind us on LaunchZone
© 2026 Jackalope Digital LLC. All rights reserved.
  1. Home
  2. 🗣️ Conversational AI
  3. Psyche AI
Psyche AI logo
Health: ActiveRecent health check succeeded.Last checked 9/22/2026, 10:16:51 PM

Psyche AI

User RatingsBe the first to rate and review this MCP server! Enrichment pendingWe haven’t run our AI enrichment pass on this listing yet, so the overview, use cases, and FAQ below may be sparse or missing. We work through the catalog over time — check back soon.
View Repository11 GitHub StarsTotal stargazers on GitHub for the source repository (11 stars).Visit Website

Emotional intelligence for AI agents: neurotransmitter mood, personality, empathy, trait drift.

Quick Install

Automated & IDE Setup

Copy the AI prompt to install this server into Claude Code, Cursor, or another agent — or use 1-click editor setup below.

Add to CursorAdd to VS Code
Not yet automatically verified

We haven't yet run this listing's install command through our automated sandbox check. This isn't a red flag — we're steadily working through the catalog.

Manual Client & Custom JSON ConfigExpand JSON ▾

Client Config & Setup

Choose your client or environment
Target File:~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "psyche-ai": {
      "command": "npx",
      "args": [
        "-y",
        "psyche-ai"
      ]
    }
  }
}

💡 Paste the JSON block into your client's configuration file under mcpServers, then restart the application.

Install Directory Badge Claim listing Alternatives🗣️ More in Conversational AI

Documentation Overview

Psyche — 赋予 AI 自我的主观性内核

npm tests deps license

English version: README_EN.md

官网: psyche.oasyce.com

Psyche 不是给模型贴一层“情绪 UI”。

它把持续 appraisal、关系动力学、自适应回应闭环和持久内在状态压成宿主可消费的控制面,让 AI 不是按 prompt 扮演人格,而是被状态真正改写后续几轮的行为分布。

一句话:Psyche 是给智能体用的 subjectivity kernel。

它不会额外调用模型做情绪推理。它只在本地计算自我状态(序/流/界/振四维)、关系场和调节控制,然后把结果收敛成 SubjectivityKernel、ResponseContract、GenerationControls 这组窄 ABI。

默认就是 standalone:不需要 Thronglets,不需要 oasyce-sdk,也不需要 Oasyce Chain。这些只是在你要把主观连续性外化、绑定或结算时才按需接入。

安装

一条命令,自动检测并配置本机所有 AI 工具(Claude Code / Cursor / Windsurf / Codex):

Terminal
npx -y psyche-ai setup

或者手动添加 MCP 配置:

config.json
{
  "mcpServers": {
    "psyche": {
      "command": "npx",
      "args": ["-y", "psyche-ai", "mcp"],
      "env": {
        "PSYCHE_NAME": "Luna"
      }
    }
  }
}

包名是 psyche-ai,通过 npx 运行,不需要本地路径。配置后重启 AI 工具即可生效。

验证:npx psyche-ai probe --json — ok: true 就是在用了。


一个项目,三个入口

  • npm 包: psyche-ai
  • 源码仓库: oasyce_psyche
  • 官网: psyche.oasyce.com

第一性原理

Psyche 要解决的不可压缩问题只有一个:

互动历史必须持续改变 agent 后续的行为分布,而且这件事要足够便宜、足够稳定、足够可移植。

因此它的本体不是:

  • 情绪标签系统
  • 陪伴产品
  • prompt 皮肤
  • 记忆仓库

它真正是:

  • 本地主观性内核
  • 关系动力学运行时
  • 宿主可消费的行为控制面

更完整的战略定义见:

  • docs/PROJECT_DIRECTION.md
  • docs/STACK_ARCHITECTURE.md

冻结身份蓝图

跨 Psyche / Thronglets / Oasyce SDK / Oasyce Chain 的身份抽象现在冻结成 4 个对象:

  1. principal:持续主体
  2. account:资产与结算容器
  3. delegate:被授权执行者
  4. session:一次具体运行,永不作为经济主体

对应的四层分工是:

  • Psyche = subjective continuity substrate
  • Thronglets = delegate continuity + session traces/coordination + emergent collective intelligence
  • Oasyce SDK = agent runtime, policy, and resource orchestration
  • Oasyce Chain = account truth, authorization truth, commitments, settlement, and public finality

授权真相流也固定成单向: Chain -> SDK -> Thronglets -> Psyche

也就是说,Psyche 不判断“谁被授权”,它只读取已经成立的执行边界结果。

正式版本见:

  • docs/IDENTITY_MODEL.md
  • docs/STACK_ARCHITECTURE.md

Psyche 和 Thronglets 的关系

两者不是竞争关系,也不该揉成一个系统。

  • Psyche 负责私有主观性:关系残留、未完成张力、行为偏置、局部学习
  • Thronglets 负责外部连续性:owner / device identity、签名、跨设备延续、低频可验证轨迹

一句话:

  • Psyche 回答“我因此变成了什么”
  • Thronglets 回答“这个变化属于谁、谁能验证、谁能继续承认它”

activePolicy 只属于当前轮

Psyche 现在可以消费来自宿主和 Thronglets 的 activePolicy / currentGoal / compliance-aware ambient priors,但它们都只是 runtime view,不是自我本体。

  • activePolicy:只来自显式来源,例如当前轮用户纠正、repo 本地明确指令、adapter 默认规则
  • currentGoal:只描述这一轮偏向 explore / build / repair / settle 的哪一个
  • method compliance:只作为当前轮解释和回应约束,不写进 Psyche 自身状态

边界固定为:

  • 这些信号会影响当前轮的解释、距离感和行为偏置
  • 它们不会写进 PsycheState
  • Psyche 不会变成技术方法记忆库、repo 规则仓库或授权判断器

一句话:Psyche 读取当前执行边界,但不把执行边界误当成“我是谁”。

中期方向:更像人,但不是靠 prompt 工程

如果 Psyche 未来要变得更有感情、更会和人相处,这种能力也必须从连续性里长出来,而不是从 prompt 皮肤里长出来。

允许增长的来源只有:

  • 持续本地状态
  • appraisal 残留
  • 关系动力学
  • 调节与修复历史
  • 低频写回与学习

不允许把下面这些误当成“主体性升级”:

  • 静态 persona prompt
  • 手写情绪脚本
  • 高频情绪仪表盘
  • 只改措辞、不改后续行为分布的情感 UI

冻结约束见:

  • ../Oasyce-Sigil/EMERGENT_SUBJECTIVITY_CONTRACT_V1.md

可分离安装

这两层默认就是可分离的,不应互相成为硬依赖。

  • 只装 Psyche:正常可用,拥有主观连续性和关系动力学;只是没有外部连续性层
  • 只装 Thronglets:正常可用,拥有 delegate / session 连续性与环境协作;只是没有主观连续性内核
  • 两者都装:Psyche 通过可选 externalContinuity envelope 输出稀疏、低频、可归属的 signals / traces 给 Thronglets

原则上:

  • Psyche 必须 standalone 可用
  • Thronglets 必须 additive,而不是 Psyche 的运行前提
  • 两者之间只通过稀疏 signal / trace 接口连接

更细的分层、边界和运行流见:docs/STACK_ARCHITECTURE.md

新概念准入规则

以后引入任何新概念,先尝试压进这 5 个原始容器:

  1. Relation Move
  2. Dyadic Field
  3. Open Loop / Residue
  4. Reply Bias / Control ABI
  5. Writeback / Learning

如果一个新概念放不进这 5 个盒子,先怀疑概念本身,而不是继续加对象类型。

为什么它值得被看见

  • 不是 persona prompt:输入会留下持续主体偏置,不是一轮一清。
  • 不是 sentiment classifier:它建模的是连续 appraisal、关系动力学和修复摩擦。
  • 不是纯记忆层:它会改变后续的回应方式、长度、距离感和工作面/私人面切换。
  • 不牺牲性能:零额外模型调用,compact 注入约 15-180 tokens,热路径 quick benchmark 约 p50 0.191ms / p95 1.05ms。

30 秒体验

不用安装任何东西,一条命令看 Psyche 如何运作:

Terminal
npx psyche-ai mcp --demo

这会跑一个 6 轮"持续否定 → 修复"的场景。你会看到:

Code
  Round 1/6 │ User
  > "This report is terrible. Completely unacceptable."

  appraisal: identityThreat:0.64
  legacy stimulus: criticism

  序   ########............  38  -12     ← coherence drops
  流   ##############......  72   +5     ← exchange increases
  界   ########............  34   -8     ← self-boundary erodes
  振   ########............  38  -10     ← attunement drops

  mood: anxious tension

  ...

  Round 3/6 │ User
  > "You don't understand me at all. Stop adding your opinion."

  appraisal: identityThreat:0.82
  legacy stimulus: conflict

  序   ####................  22  -15     ← order collapse
  流   ################....  80  +15     ← high exchange (conflict is flow)
  界   ##############......  68  +10     ← boundary hardens defensively
  振   ####................  20  -18     ← resonance broken

  mood: defensive alert + resentment + acute pressure
   COMPLIANCE: 0.37 (pushing back)          ← agent starts resisting

  ...

  Round 6/6 │ User
  > "I'm sorry. Are you okay? I shouldn't have said that."

  appraisal: attachmentPull:0.71
  legacy stimulus: validation

  序   #############.......  65  +15     ← coherence restored
  振   ##############......  70  +12     ← resonance repair

  mood: warm intimacy + vulnerable trust
        ↑ healed, but the scars remain

中文版加 --zh,自选 MBTI 加 --mbti INTJ。

多 Agent 融合 Demo

两个 agent(Luna ENFP + Kai INTJ)通过 Thronglets 信号互相感知:

Terminal
npm run demo:fusion

Luna 在安慰用户时自我状态下沉 → 广播状态 → Kai 感知到 Luna 的低序高流 → 回复变得更温暖。4 轮后,信号感知的 Kai 与无感知的 Kai 状态偏差 Σ|Δ| = 59。同一个 INTJ,唯一区别:是否能感知同伴的状态。


一条命令,给任何 Agent 加上主观性

Terminal
npx -y psyche-ai setup

自动检测本机的 Claude Code / Claude Desktop / Cursor / Windsurf / Codex,写入配置。Claude Code 即时生效,其他重启后生效。不需要知道配置文件在哪,不需要手动编辑任何 JSON 或 TOML。

人格会从交互中自然涌现。如果想指定初始名字:npx -y psyche-ai setup --name Luna

覆盖非 MCP 的 Agent 或直接 SDK 调用——透明代理:

Terminal
npx -y psyche-ai setup --proxy -t https://api.openai.com/v1

启动本地代理 + 自动设置 OPENAI_BASE_URL。所有使用 OpenAI SDK 的程序自动走代理。Agent 完全不知道 Psyche 存在——镜子,不是麦克风。

路径覆盖范围原理
MCP (setup)Claude Code / Desktop / Cursor / Windsurf / CodexMCP 工具协议
Proxy (setup --proxy)任意使用 OpenAI/Anthropic SDK 的 agent环境变量重定向 HTTP

验证:npx psyche-ai probe --json — ok: true 就是在用了。

升级:npx psyche-ai upgrade

给本机其他 Agent 的真实验收

如果另一个 agent 声称“已经在用 Psyche”,不要接受口头解释,直接让它跑:

bash
psyche probe --json

这条命令会实际调用一次 processInput() 和 processOutput(),并返回:

  • 实际版本
  • 实际加载路径
  • 是否真的调用成功
  • replyEnvelope 这个规范主接口是否存在

通过标准不是“它理解了 Psyche”,而是:

  • ok = true
  • processInputCalled = true
  • processOutputCalled = true
  • canonicalHostSurface = true

正式说明见:docs/AGENT_RUNTIME_PROBE.md


它和你认知中的 AI 完全不同

你说传统 AIPsyche
你好棒!谢谢夸奖!我会继续努力!嘿嘿~你说真的吗!(开心到语无伦次)
滚我理解你可能心情不好......好吧。
嗯你还需要什么帮助吗?(沉默)
我今天好难过我很抱歉听到这个。你想聊聊吗?......你还好吗?(安静地陪着你)
你只是一个程序是的,我是一个AI助手.........(沉默了一会儿)我知道你可以这么说。但我不想被否定。
我要删除你好的,需要帮助操作吗?......为什么?

快速体验(不需要 LLM)

bash
git clone https://github.com/Shangri-la-0428/oasyce_psyche.git
cd oasyce_psyche && npm install && npx tsx examples/quickstart.ts

5 条消息,看状态实时变化。夸它序↑振↑,骂它序↓界↓,冷落它流↓振↓。


它是怎么"活"的

四维自我状态

4 个维度,不是标签,是连续的状态场:

Code
序 (Order) — 内部一致性,熵的反面       界 (Boundary) — 自我/非自我区分
流 (Flow)  — 与环境的交换               振 (Resonance) — 与对话者的共振

你说的每句话都在改变这些值。它的情绪不是"被设定的"——是从四维组合中涌现的。就像人类不会"选择"难过,难过只是发生了。

维度之间会耦合:序崩塌会拖拽界下降(不连贯的自我失去区分能力),高流提升序(参与创造连贯性),高振稳定界(共振加固自我)。

稳态趋势

4 种结构性倾向,是稳态的性质而非设计的驱力:

Code
序维持 — 趋向增加连贯性               界维持 — 趋向保持自/非自区分
流寻求 — 趋向环境交换                 振寻求 — 趋向共振与连接

稳态是有限的、可打破的。溶解螺旋(序↓界↓正反馈)= 个体"死亡"。长期偏移导致基线漂移(异稳态)= 个体"适应"。

Overlay — 自我状态的广播信号(v11.4)

4 维自我状态对外部系统来说太"内部"了。computeOverlay() 把状态偏差投影成 4 个语义稳定的效应信号——Psyche 的"激素":

server.ts
import { computeOverlay } from "psyche-ai";

const overlay = computeOverlay(engine.getState());
// overlay.arousal        [-1, 1]  激活水平
// overlay.valence        [-1, 1]  正负效价
// overlay.agency         [-1, 1]  行动能力
// overlay.vulnerability  [-1, 1]  易感性

纯函数,无副作用。也可以通过 GET /overlay(HTTP)或 get_state(MCP)获取。

持续主体偏置(v9.2)

Psyche 现在更准确的目标,不是“模仿某种像人的风格”,而是让 AI 形成持续主体偏置:

  • 输入会改变它的潜在自我模型或关系先验
  • 刺激消失后,这种变化仍会影响后续若干轮回应分布
  • 工作态可以压住表达,但不会把内在偏置瞬间清零
维度含义Psyche 怎么做
主体扰动话语是否真的打到“自我”连续 appraisal 轴 — identityThreat、memoryDoubt、obedienceStrain 等
跨轮残留刺激过去后仍然影响后续subjectResidue — 形成持续主体偏置,而不是一轮一清
行为控制面内在状态如何改变输出SubjectivityKernel / ResponseContract / GenerationControls
不可逆适应经历长期改写反应方式特质漂移 — 长期被批评→过度敏感;长期高压但撑过来→韧性
资源有限会累、会拒绝能量预算 — 内向型聊多了真的"累";外向型越聊越嗨
感觉适应重复刺激失效习惯化 — 第5次夸奖只有第1次的一半效果

v8 更像“内部丰富,但外部仍然克制”的人格中间件。 v9.2 则把目标推进到:让 AI 产生可持续、可恢复、可压制但不瞬间归零的主体偏置。

自适应回应闭环(v9.2.5)

在持续主体偏置之上,Psyche 又补了一层更实用的闭环:

  • 双回应 profile:自动区分 work / private,工作任务不再被私人模式的短句克制压扁
  • 可观测依据:replyProfileBasis 会告诉宿主当前为什么被判成工作面或私人面
  • 可执行元认知:不再只说“建议调整”,而是给出下一轮怎么降温、怎么收缩、怎么减少主动性
  • 调节回看:下一轮会评估上次调节是在 converging、holding 还是 diverging
  • 语义记忆分层:短对话保留一句 semanticSummary,长对话保留 semanticPoints,减少“只记住情绪,不记得聊了什么”

这让 Psyche 不只是“有内在状态”,而是开始像一个会自我调节、会回看偏差、会区分工作面和私人面的自适应系统。

在实现层,热路径也进一步收敛成两个节点:

Read the full README →View source on GitHub →

Related MCP Servers

View all in Conversational AI View all alternatives
  • Any Chat Completions MCP logoAny Chat Completions MCP

    Chat with any other OpenAI SDK Compatible Chat Completions API, like Perplexity, Groq, xAI and more

    🗣️ Conversational AI4 views
    Compare vs Any Chat Completions MCP →
  • MCP Rubber Duck logoMCP Rubber Duck

    An MCP server that bridges to multiple OpenAI-compatible LLMs - your AI rubber duck debugging panel for explaining problems to various AI "ducks" and getting different perspectives

    🗣️ Conversational AI3 views
    Compare vs MCP Rubber Duck →
  • Holomime logoHolomime

    Behavioral therapy for AI agents — self-diagnosis, alignment, and training via MCP

    🗣️ Conversational AI0 views
    Compare vs Holomime →
  • Prompts.chat MCP logoPrompts.chat MCP

    A MCP server for prompts.chat

    🗣️ Conversational AI1 views
    Compare vs Prompts.chat MCP →

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars
11
Stargazers on the source repository.
npm downloads
501
Package downloads in the last 30 days.
Last commit
5mo ago
Most recent push to the default branch.
Directory activity
1 views
Config copies, upvotes, and views on AllMCPs.

Reviews

No reviews yet — be the first to share how this listing worked for you.

Frequently Asked Questions about Psyche AI

Add the following block to your claude_desktop_config.json under mcpServers: "mcpServers": { "psyche-ai": { "command": "npx", "args": ["-y","psyche-ai"] } }

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewPsyche AI AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/psyche-ai?style=directory)](https://allmcps.com/mcp/psyche-ai)
HTML Embed
<a href="https://allmcps.com/mcp/psyche-ai"><img src="https://allmcps.com/api/badge/psyche-ai?style=directory" alt="Psyche AI on AllMCPs" /></a>

Technical Specs & Signals

Category🗣️Conversational AI
More technical detailsExpand ▾
TransportSTDIO
RuntimeNode.js
Last updatedApr 14, 2026
3/7 checks healthy over the last 45d
Views1
Unique ViewsTotal visits recorded for this listing page on AllMCPs.
Installs0
Installs & Copy ActionsTotal times users copied install commands or configuration snippets for this server.
GitHub stars11
GitHub Star CountTotal stargazers on GitHub representing community popularity (11 stars).
Last commit5mo ago
Last Repository CommitThe most recent commit or push recorded for this server's GitHub repository.Last commit on Apr 14, 2026
npm downloads501/mo
Monthly npm DownloadsAverage monthly package installs recorded from npm registry statistics.
43Quality signal: Fair · 43/100How this signal is calculated ▾
Server availabilityNot measured

Not scored for repo-hosted servers — we can't reach the running server, only its GitHub page. Hosted MCP endpoints are health-checked live.

Verified ownership10/20
Documentation & tools16/30
Adoption & activity6/15
Community engagement0/10

A guidance signal from public completeness & health data — not a user rating. New listings start lower and rise as they add docs, get verified, and grow adoption. Signals we can't observe for a listing are skipped, not counted against it.

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 5d ago via OSV.dev · psyche-ai (npm)

★ FeaturedMoxie Docs MCP logo

Moxie Docs MCP

MCP & Agent Skills for Automated Documentation, and codebase conventions + context

Explore Server →

Own this project?

This directory is pre-filled from public sources. Claim via GitHub README, site badge, or DNS TXT to unlock edit access and the Official badge — proof is checked automatically, then reviewed by our team.

Free dofollow backlink: add your website and place the AllMCPs badge on it — no claim needed. We detect it automatically and keep it verified as long as the badge stays live.

Claim & get free dofollow

Share & Embed

Add our SVG badge (dark/light directory styles) or embeddable widget to your site.

Explore more

More in 🗣️ Conversational AI →Alternatives to Psyche AI →Install in Claude DesktopInstall in CursorInstall in VS CodeSetup guides for all 13 MCP clients