MyAITwin

com.lutolearn由 lutolearn.com 发布官方发布者v1.0.0更新于 Oct 26, 2026

A personal RAG database you build from chat, so AI creates work that sounds like you.

已验证Streamable HTTP可网页运行数据库知识与记忆

概览

AI 生成的概览

个人 RAG 数据库与语义搜索服务，让助手保存你的知识与表达风格，并在引用来源的前提下检索使用。

可以做什么
  • 存储知识条目、语音笔记、长文档、抓取的网页以及创作记录
  • 在所有知识中做语义搜索，或为创作任务分别检索技能与知识
  • 按类型、按标签检索条目，或列出最近新增内容
  • 查看结构与来源清单，发现重复出现的模式，并就某主题综合多条知识
  • 管理个人知识库：初始化会话、更新条目、定义或修改知识类型，以及删除条目
适用场景
适合希望助手在不同对话中产出带有你个人素材与语气的内容，例如依据会议记录和你存储的邮件风格起草跟进邮件。也适合愿意在聊天中逐步搭建并维护个人知识库的人。
运行要求
需要连接服务商的远程 MCP 端点（Streamable HTTP 的 /mcp 地址），并通过 OAuth 2.1（PKCE、动态客户端注册）登录。需用邮箱和魔法链接注册账号，并使用支持 MCP 的客户端；README 提到 Claude Pro、Claude Team 与 ChatGPT Pro。未声明环境变量或请求头。
安装前请注意
存储的内容会发送给并由服务商保存，README 称其使用 Supabase 与 Pinecone 作为存储层，并声明数据不用于训练模型、不与第三方共享。多数工具会写入数据，delete_knowledge 为破坏性操作。账号删除被描述为立即且不可恢复。服务仅在早期访问阶段免费，条款可能变化。

安装

在 SourceWeft 中

  1. 打开 控制台中的 MyAITwin，将其添加到工作区。
  2. 连接 所需的 凭据。它们会被私密存储，绝不会在公开页面上展示。
  3. 为需要使用其工具的对话启用该服务。

Web executable，通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体，或 SourceWeft 对话中。智能体会读取该服务器的安装指南，先向你展示发布者、端点和登录方式，经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/com.lutolearn%2Fmyaitwin/install.md，并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前，先告诉我它的发布者、端点和登录方式，等我确认。不要把 API 密钥写进命令或我会提交的文件；问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。 客户端会在首次连接时完成 OAuth 登录。

{
  "mcpServers": {
    "myaitwin": {
      "type": "http",
      "url": "https://myaitwin.lutolearn.com/mcp"
    }
  }
}

README

MyAITwin MCP

A personal RAG database and semantic search engine you build and control from inside your AI chat. Store your knowledge, voice, and skills as you work. Retrieve them in seconds, source always cited. Your AI then creates output that is recognisably you, in every conversation.

Live at https://myaitwin.lutolearn.com. Free during early access.

What it is

MyAITwin is two things at once.

The toolbox. A production-grade RAG database with semantic search that you shape from chat. You define the structure, the types, the tags. It is yours, it is visible, and you are the architect of it.

The twin. The layer on top that greets you, guides you, assesses what you store, and creates output that sounds like you. It knows the difference between what you know (your knowledge) and how you say things (your skills), and it uses both.

Install

Three steps. Under two minutes.

  1. Sign up at https://myaitwin.lutolearn.com/ with your email.
  2. Click the magic link, then copy your personal MCP URL from /create.
  3. In Claude Desktop: Settings → Connectors → Add custom connector → paste your URL.

Or use the canonical OAuth-authenticated endpoint:

  • URL: https://myaitwin.lutolearn.com/mcp
  • Transport: Streamable HTTP
  • Auth: OAuth 2.1 with PKCE (S256) and Dynamic Client Registration

Requires a client with MCP capability. Currently Claude Pro, Claude Team, and ChatGPT Pro.

The 19 tools

Storing knowledge

ToolWhat it does
add_knowledgeStore a typed, tagged knowledge item
add_voice_noteStore a voice note transcript with automatic extraction
add_documentStore a long document with automatic chunking
add_from_urlFetch and store a web page
add_reference_recordStore a creation event linking knowledge and skills used

Retrieving knowledge

ToolWhat it does
search_twinSemantic search across all knowledge
search_for_creationDual search returning skills and knowledge separately
get_by_typeRetrieve all items of a specific type
get_by_tagRetrieve all items with a specific tag
list_recentList recently added items

Understanding your twin

ToolWhat it does
get_schemaOverview of your types and how many items you have
get_sourcesList all source documents
find_patternsSurface recurring patterns across your knowledge
synthesiseSynthesise across multiple knowledge items on a topic

Managing your twin

ToolWhat it does
get_welcomeSession initialisation and system prompt
update_knowledgeUpdate an existing item
add_schema_typeDefine a new knowledge type
update_schema_typeUpdate an existing type definition
delete_knowledgeDelete an item (destructive)

All tools are annotated with title, readOnlyHint, and destructiveHint per the MCP spec. Of the 19: 10 read-only, 8 write (non-destructive), 1 destructive (delete_knowledge).

How it works

RAG is Retrieval-Augmented Generation. It is the architecture that lets AI answer using your specific knowledge rather than its training data alone.

Two layers:

  • Supabase (PostgreSQL) for structured records with types, tags, and provenance.
  • Pinecone for vector embeddings, so you can search by meaning rather than exact words.

When you search, both layers work together and return results ranked by relevance. Every result is cited with source and date, and tagged with provenance: personal (your own thinking), organisational (from your organisation), or external (from someone else).

The architectural insight worth getting right:

Knowledge is what you know. Facts, decisions, transcripts, observations.

Skills are how you express things. Your LinkedIn voice. Your email style. Your proposal structure.

Exceptional output needs both. Take a meeting transcript and ask for a follow-up email. The twin needs the transcript and your email skill to produce something that is accurate and unmistakably yours. Neither alone is enough.

Security and privacy

  • Bearer token authentication on every request, hashed at rest.
  • OAuth 2.1 with PKCE for connector-style integration. No shared secrets, no static credentials.
  • Multi-tenant data isolation: each user lives in their own namespace. Other users can never read your data. Verified by a 35-check cross-tenant test suite.
  • Rate limiting per tenant.
  • Append-only audit log on every tool call.
  • Prompt injection guardrails on stored content.
  • Your data is used only to provide the service. Never used to train AI models. Never shared with third parties.
  • You can delete your account and all data instantly from /create. Deletion is immediate and irreversible.

Privacy policy: https://myaitwin.lutolearn.com/privacy Security contact: [email protected] Privacy contact: [email protected]

Distribution

  • Official MCP Registry: com.lutolearn/myaitwin
  • Anthropic Connectors Directory: submitted, in review
  • Listed at: Glama, mcp.so, mcp.directory, mcpserverfinder, Hugging Face, awesome-mcp-servers

License

MIT. See LICENSE.

Links

MyAITwin MCP by Luto.

来源：README.md，提交 2f1ce3d

工具

0
该服务器需要登录，因此 SourceWeft 无法在没有账号的情况下读取其工具，也未收录。已连接账号所能看到的内容可能因账号及其权限而异。

版本历史

1
  1. v1.0.0最新Sep 16, 2026

评分

已满足适用检查项的 75%

该评分衡量此收录条目的完整程度，依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价，也不表示服务器安全或当前可用：可连通性是带日期的检查结果，评分也不包含安装量或用户评价。

  • 近 30 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list，或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分

    请在仓库或注册中心条目中声明许可证。