Claudesidian MCP 插件 for Obsidian

by ProfSynapse

Model Context Protocol (MCP) 集成插件,适用于 Obsidian,允许 AI 助手通过原子操作和结构化的记忆系统与您的知识库进行交互。需要按照安装步骤将插件文件放入 Obsidian 的插件文件夹中。

Developer toolsstdioCommunity

Repository-wide counts · Cached 2026-03-07

Overview

The Claudesidian MCP 插件 for Obsidian MCP server is a publicly available project. Review the upstream repository for installation instructions, supported tools, compatibility, permissions, and current maintenance status.

Configuration

Configuration, transport, authentication, and runtime requirements vary by project. Open the repository before connecting and use the smallest set of credentials and permissions required.

Open the Claudesidian MCP 插件 for Obsidian repository to read the latest documentation.

KEEP EXPLORING

Compare source, connection, and authentication details before choosing an implementation.

View the complete category

模型上下文协议服务器

modelcontextprotocol

Community

一组用于模型上下文协议(MCP)的参考实现,展示了对大型语言模型(LLM)工具和数据源的安全且受控的访问方式。

Context7 Platform - Up-to-date Code Docs For Any Prompt

upstash

Community

Context7 MCP server providing up-to-date, version-specific documentation and code examples for libraries, enabling coding agents to fetch accurate docs and code snippets. Requires an API key for higher rate limits, passed via CONTEXT7_API_KEY header.

Playwright MCP

Microsoft Corporation

Community

A Model Context Protocol (MCP) server that provides browser automation capabilities using Playwright. Enables LLMs to interact with web pages through structured accessibility snapshots, bypassing the need for screenshots or visually-tuned models.

AIHawk

feder-cr

Community

AIHawk is an anti detect browser and web browsing agent, open source, with an MCP server for coding agents: undetected, no captchas, no blocks. It requires an OpenRouter API key for the standalone web UI mode, which can be provided via the --openrouter-key flag or the OPENROUTER_API_KEY environment variable or a .env file in the running directory.

FROM THE SOURCE

Repository README

Build-time snapshot · Retrieved 2026-10-05

View original

MCP Toplist

Nexus Obsidian Banner

Release License Obsidian Node

Nexus MCP for Obsidian

Nexus gives AI agents and built-in chat access to your Obsidian vault so you can read, write, search, organize, and automate notes in natural language while keeping storage local to the vault.

Nexus can be used in two ways:

  • Inside Obsidian with native chat (hook up to your favorite provider or agentic platform!)
  • From external agents like Claude Desktop, Claude Code, Codex CLI, Gemini CLI, Cursor, Cline, and other MCP clients

Nexus is the successor to Claudesidian. Legacy installs in .obsidian/plugins/claudesidian-mcp/ still work.

Setup

  • Install the latest release from GitHub Releases: manifest.json, styles.css, and main.js
  • Put them in .obsidian/plugins/nexus/ and enable Nexus in Obsidian
  • Native chat in Obsidian: Provider setup and Native chat guide for chat, live voice, read-aloud, and built-in audio/video generation
  • External agent over MCP: use Nexus settings -> Get started -> External agents to create connector.js and update Claude Desktop, then see MCP setup guide and Recommended system prompt
  • External agent over CLI (no MCP config): install the nexus command from Get started -> External agents -> Local CLI so shell agents (Claude Code, Cursor, Codex) can drive your vault directly — see the Nexus CLI guide. Installation is user-triggered, writes only outside the vault, and on Windows adds the Nexus directory to the current user's PATH without administrator rights; uninstall removes only entries Nexus recorded as its own.
  • Optional desktop features: Semantic search, Adaptive search, and Apps and integrations

Native chat works on desktop and mobile. MCP clients, local desktop providers, and semantic search are desktop-only.

Mobile Support (Experimental)

Native chat works on mobile (iOS and Android). Desktop-only features gracefully skip loading on mobile -- they will not appear or cause errors.

Feature Mobile Desktop
Native chat Yes Yes
Workspace memory and tasks Yes Yes
Skills (author and load) Yes Yes
MCP clients (Claude Desktop, Cursor, etc.) No Yes
Local CLI bridge (nexus command) No Yes
Semantic search (local embeddings) No Yes
Ingestion (PDF, audio, DOCX) No Yes
Web capture to Markdown (web capture-markdown) Yes Yes
Composer and the rest of Web Tools No Yes
Data Analysis (Python over CSV/Excel) No Yes

Mobile support is new and may have bugs. Please report issues on GitHub.

Network Use

Nexus does not call out on its own. Nothing is sent when the plugin loads, and there is no telemetry, analytics, crash reporting, or usage tracking of any kind. Your notes stay in your vault unless you do something that sends them somewhere. Every request below is listed with the action that causes it.

AI provider APIs. Contacted only once you enter that provider's API key and then use it. Your prompt — and whatever vault content the agent decides to include — goes to the provider you chose, and your key is only ever sent to the provider it belongs to.

Chat and models openrouter.ai, api.openai.com, api.anthropic.com, generativelanguage.googleapis.com, api.mistral.ai, api.groq.com, api.deepseek.com, api.perplexity.ai, router.requesty.ai, api.githubcopilot.com
Voice and transcription api.elevenlabs.io, api.deepgram.com, api.assemblyai.com, streaming.assemblyai.com
Sign-in flows auth.openai.com and chatgpt.com (OpenAI Codex), api.github.com (GitHub Copilot)

Ollama and LM Studio talk to your own machine and leave it entirely.

Downloads for optional desktop features. A few features fetch their engine the first time you enable them, rather than shipping inside main.js — the download is cached and does not repeat. Never enabling the feature means never making the request.

Feature Downloads from
Semantic search cdn.jsdelivr.net (transformers.js), huggingface.co (embedding model)
Vector cache cdn.jsdelivr.net or unpkg.com (sqlite3.wasm)
Data analysis cdn.jsdelivr.net (Pyodide), files.pythonhosted.org (openpyxl)
PDF ingestion cdn.jsdelivr.net (pdf.js worker)
In-browser local models cdn.jsdelivr.net, esm.run, huggingface.co, raw.githubusercontent.com (model weights)

Housekeeping. Opening Nexus settings checks for a newer version (api.github.com, raw.githubusercontent.com). Asking an agent to load the built-in guides workspace fetches the current guides (raw.githubusercontent.com).

Not requests. Two things look like external hosts but aren't. Calls to OpenRouter and Requesty carry HTTP-Referer: https://synapticlabs.ai and X-Title: Nexus — the app-attribution headers those routers show in their dashboards. They are header values; nothing is fetched from that address. And buttons like Download Node.js or Get an API key open nodejs.org, ollama.com, lmstudio.ai, claude.ai, github.com, or the provider's own console (platform.openai.com, console.anthropic.com, aistudio.google.com, and so on) in your browser — the plugin never retrieves them.

Use Cases

If you want to... Start here
Connect Claude Desktop, Codex CLI, Gemini CLI, Cursor, Cline, or another MCP client MCP setup
Drive your vault from the shell with Claude Code, Cursor, or Codex — no MCP config Nexus CLI
Configure built-in chat providers inside Obsidian Provider setup
Use live voice, read notes aloud, or save spoken audio back into your notes Native chat
Generate voice audio or text-to-video files directly into your vault Native chat
Give your agent better instructions for using Nexus Recommended system prompt
Manage long-running work with persistent workspace context Workspace memory
Track projects, tasks, blockers, and dependencies Task management
Search notes and past conversations by meaning Semantic search
Have search learn from the notes you actually open, fully on-device Adaptive search
Edit selected text directly in notes Inline editing
Open webpages in Obsidian and save them as Markdown, PNG, or PDF (experimental) Apps
Convert PDFs and audio files to Markdown notes — right-click in vault or auto on add (experimental) Apps
Merge PDFs, concat markdown, or mix audio tracks into one file (experimental) Apps
Author and load reusable agent Skills straight from your vault Apps
Run Python/pandas analysis over your vault's CSV and Excel data (experimental, desktop) Apps
Create recurring routines and reusable workflows Workflow examples
Understand the MCP design and available tools Two-tool architecture
Extend Nexus with optional apps Apps

Prompt For Your Agent

If you want another agent to walk you through setup, paste this:

Help me set up Nexus for Obsidian and guide me step by step.

Use these docs as the source of truth:
- README: https://github.com/ProfSynapse/claudesidian-mcp/blob/main/README.md
- Provider setup: https://github.com/ProfSynapse/claudesidian-mcp/blob/main/guide/provider-setup.md
- MCP setup: https://github.com/ProfSynapse/claudesidian-mcp/blob/main/guide/mcp-setup.md
- Recommended system prompt: https://github.com/ProfSynapse/claudesidian-mcp/blob/main/guide/recommended-system-prompt.md
- Native chat guide: https://github.com/ProfSynapse/claudesidian-mcp/blob/main/guide/native-chat.md

Start by figuring out whether I want native chat inside Obsidian, an external MCP agent, or both. Ask for my OS and the agent I want to use if that matters. Then walk me through the exact setup path, one step at a time.

When a config file needs to be edited, show the exact snippet with my vault path inserted. Do not invent config formats or skip restart/reload steps. If multiple setup paths are possible, recommend the simplest one first.

Development

npm install
npm run dev
npm run build
npm run test
npm run lint

License

MIT. See LICENSE.