xiaofenshen-mcp-server
Click on "Deploy 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., "@xiaofenshen-mcp-server What image prompt templates are available?"
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.
小分身公开 MCP
这是 小分身(TwinWrite)的公开 MCP 仓库,把已经免登录的只读接口转成 MCP 工具,给 Cursor、Claude Desktop 等外部 Agent 使用。
小分身是什么
小分身 是一个通过对话、长期记忆和「造梦」持续进化的个人 AI 创作分身。官网:xiaofenshen.com。
你通过对话与训练告诉它你的语气、句式与偏好,它再用你的口吻帮你写下一篇。你可以创建多个分身,各自专注不同的创作场景;随着对话、记忆与反馈不断积累,它对你的理解会越来越深,输出也越来越像你本人。
常见用法是模仿你的风格写公众号、小红书等自媒体内容:在对话里粘贴代表作或直接训练,让分身学你的表达,再把 AI 写出来的稿子改得不像 AI、更像你本人。
本仓库是公开的薄客户端:只保存工具契约和 HTTP 转发。业务数据与实现都在小分身线上服务(xiaofenshen.com),这里不复制抓取、计费或分身逻辑。
Related MCP server: @prosodyai/mcp-docs
当前开放
工具 | 上游 |
|
|
|
|
|
|
这三条产品侧本来就不挂登录。MCP 不新增鉴权,也不扩大可读范围。
明确不做(本版本)
热点、网页/微信抓取、分身对话、记忆、生图、公众号搜号等,都要用户身份或付费凭据。后续会走用户 token 授权,再在小分身加稳定 API,本仓库只增加对应工具名和转发,不把实现搬过来。
使用
需要 Node 22+。
pnpm install
pnpm test
pnpm buildCursor mcp.json 示例:
{
"mcpServers": {
"xiaofenshen": {
"command": "npx",
"args": ["-y", "github:TwinWrite/xiaofenshen-mcp-server"]
}
}
}环境变量 | 默认 |
|
|
|
|
相关链接
Available Tools
3 toolsget_product_docsA
读取小分身公开帮助文档。不传 slug 返回 /llms.txt 索引;传 slug 返回对应 Markdown。
Inputs: slug.
list_image_prompt_templatesA
列出小分身启用的生图提示词模板(系统级只读,含 prompt 与效果图)。可选 slug 过滤单条。不会触发生图。
Inputs: slug.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 3 tools
Each tool serves a completely distinct purpose: reading public shares, listing image prompt templates, and fetching product docs. There is no overlap or ambiguity between them.
All tool names follow a consistent verb_noun pattern with snake_case: read_public_share, list_image_prompt_templates, get_product_docs. The naming is predictable and uniform.
With only 3 tools, the server is tightly scoped for its read-only informational purpose. Each tool earns its place and covers a distinct resource, fitting comfortably within the ideal range.
The server covers the three main public-facing resource types (shares, templates, docs) with read operations. No obvious gaps exist for the stated read-only domain, as operations align with the server's purpose.
Maintenance
Related MCP Connectors
Read-only MCP tools for AI agent discovery, structured resources, and NIULAI information.
Read-only public services, cases, resources and FAQ from Chenfeng AI.
Free public MCP for AI agents — 193 tools, 44 workflows. No API key.
Read-only search of 2,189 AI agent tools (MCP servers, x402 APIs) with daily automated trust scores.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceExposes OpenAPI specifications as MCP tools, enabling AI assistants to explore and understand API structures, endpoints, schemas, and documentation through semantic queries.19 npmMIT

@prosodyai/mcp-docsofficial
AlicenseAqualityBmaintenanceExposes ProsodyAI documentation, SDK references, REST API reference (OpenAPI), and curated implementation recipes to AI coding agents via MCP tools and resources.8MIT- AlicenseNot gradedqualityDmaintenanceProvides MCP tools to list and search OpenAI Agents SDK documentation, enabling LLMs to retrieve documentation topics and content via natural language queries.78 PyPIMIT
- FlicenseNot gradedqualityDmaintenanceEnables agents to query real-time OpenAPI documentation of backend services, providing tools to list services, endpoints, and schemas via MCP.-