SiYuan Sisyphus MCP
Enables AI agents to safely read, search, edit, and organize SiYuan notes, with git-like document timeline and notebook-level permissions.
README
SiYuan Sisyphus MCP & CLI
<p align="left"> <a href="https://www.npmjs.com/package/siyuan-sisyphus"> <img src="https://img.shields.io/npm/v/siyuan-sisyphus?style=flat-square&color=%23cb3837" alt="npm version"> </a> <a href="https://github.com/yangtaihong59/siyuan-plugins-mcp-sisyphus/blob/main/LICENSE"> <img src="https://img.shields.io/github/license/yangtaihong59/siyuan-plugins-mcp-sisyphus?style=flat-square&color=%23007ec6" alt="license"> </a> <a href="https://yangtaihong59.github.io/siyuan-plugins-mcp-sisyphus/"> <img src="https://img.shields.io/badge/docs-VitePress-646cff?style=flat-square&logo=vitepress" alt="Documentation"> </a> <a href="https://github.com/yangtaihong59/siyuan-plugins-mcp-sisyphus/releases"> <img src="https://img.shields.io/github/v/release/yangtaihong59/siyuan-plugins-mcp-sisyphus?style=flat-square&color=%23007ec6" alt="GitHub release"> </a> </p>
<p align="left"> <a href="https://github.com/yangtaihong59/siyuan-plugins-mcp-sisyphus/blob/main/README.md">English</a> | <a href="https://github.com/yangtaihong59/siyuan-plugins-mcp-sisyphus/blob/main/README_zh_CN.md">中文</a> | <a href="https://yangtaihong59.github.io/siyuan-plugins-mcp-sisyphus/">Documentation</a> </p>
Latest:
v0.4.11— Adds sponsor acknowledgements to thank recent supporters. Document Timeline can still be disabled cleanly from settings.
<p align="center"> <img src="docs/archive/timeline.png" alt="Document Timeline" width="720"> </p> <p align="center"><em>Document Timeline: named snapshots, visual diff, and rollback for SiYuan notes.</em></p>
What It Is
SiYuan Sisyphus lets AI agents safely read, search, edit, and organize your SiYuan workspace.
It ships as both a SiYuan plugin and a standalone CLI:
- MCP plugin: connect SiYuan to Claude Desktop, Claude Code, Codex, Cursor, Cherry Studio, Cline, and other MCP-capable clients.
- CLI
siyuan-sisyphus: let agents, terminals, and scripts call SiYuan directly through short commands.
Both entry points use the same permission model and the same underlying SiYuan operations.
Why It Is Different
- Git-like document timeline: create named timeline nodes for a document, compare historical snapshots with the current state, and roll back when needed.
- AI-friendly note access: the
fstool lets agents work with human-readable paths such as/Notebook/Project/Note, hiding block IDs and document-tree details when they are not needed. - MCP and CLI together: use MCP for multi-step agent workflows, or CLI for lightweight terminal and script automation.
- Notebook-level safety: give each notebook its own access level:
none,r,rw, orrwd. - Low-context tool design: 100+ SiYuan capabilities are grouped into 12 action-routed tools, with detailed help available only when the agent asks for it.
- Practical connection setup: the plugin settings page provides copy-ready connection snippets for common AI clients and deployment styles.
Git-Like Document Timeline
The timeline feature gives ordinary SiYuan documents a source-control-style safety layer:
- create named timeline nodes for the current document;
- compare a historical snapshot with the current document;
- switch between unified diff and split diff;
- use a minimap-style change navigator and collapse unchanged blocks;
- roll back the whole document, or restore supported parsed blocks individually.
This is a document-focused timeline built on SiYuan snapshots. It is intentionally not a full Git replacement or a complete source-control workflow.
MCP And CLI Entry Points
Use MCP when an AI client should discover tools and combine them across a longer task. This is the best fit for agent workflows that need to search, read, edit, inspect databases, and verify results.
Use CLI when a terminal command is enough. It avoids long tool schemas in the model context and works well for scripts, automation, and small one-shot tasks.
npm i -g siyuan-sisyphus
siyuan-sisyphus init
siyuan notebook list
Safety Model
SiYuan Sisyphus is designed around explicit user control:
- each notebook can be read-only, writable, deletable, or hidden from AI;
- dangerous actions such as delete, move, replace, and asset upload are treated separately;
- MCP and CLI share the same core behavior, so switching entry points does not create a second permission model;
- remote and Docker use cases go through the SiYuan HTTP API instead of assuming direct access to your local workspace files.
Quick Start
- Install the plugin from the SiYuan marketplace, or follow the source installation guide.
- Open
Plugin -> SiYuan Sisyphus MCP & CLI -> Settings. - Choose MCP or CLI in the connection page.
- Copy the generated client configuration, or initialize the CLI with
siyuan-sisyphus init. - Verify with a read-only task such as listing notebooks or reading the SiYuan version.
For full setup steps, use the documentation links below.
Read The Docs
Support
If you find this project helpful, please consider supporting it! Your support is what keeps me motivated to maintain and improve it.
Sponsor Thanks
Thanks to undefined, Fngd Z, and all other kind supporters for sponsoring this project.
<p align="left"> <img src="docs/archive/thank.jpeg" alt="Support QR code" width="280"> </p>
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.