aaronsb/obsidian-mcp-plugin

High-performance Model Context Protocol (MCP) server for Obsidian that provides AI tools with direct vault access through semantic operations and HTTP transport.

What it solves

This plugin allows AI assistants to interact directly with an Obsidian vault. Instead of providing a single document as context, it enables AI to read, write, search, and navigate the connected structure of your notes, including links, tags, and backlinks, treating the vault as a connected knowledge graph rather than a collection of isolated files.

How it works

It implements a Model Context Protocol (MCP) server that runs directly inside Obsidian as a plugin. This removes the need for external Node processes or separate API bridges. AI clients (such as Claude Desktop, Claude Code, or Continue.dev) connect to this server via HTTP or HTTPS to execute a set of tools that can perform file operations, graph traversal, and database queries.

Who it’s for

Obsidian users who want to use AI assistants to synthesize information across their personal knowledge base, discover connections between notes, or automate the organization of their vault.

Highlights

  • Graph Intelligence: Supports multi-hop traversal, path finding between concepts, and backlink analysis to help AI understand the relationship between notes.
  • Detailed Toolset: Provides 8 tool families covering vault operations (list, read, create, search), content editing, graph navigation, and system info.
  • Plugin Integrations: First-class support for executing Dataview queries and managing Obsidian Bases database views.
  • Granular Security: Includes API key protection, read-only mode, and path allow/block lists to control exactly what the AI can access and modify.
  • Easy Setup: Offers a .mcpb bundle for one-click installation into compatible AI clients like Claude Desktop.

Related

  • Project
    MarkusPfundstein/mcp-obsidian

    An MCP server that allows AI assistants to interact with Obsidian notes via the Local REST API plugin, enabling reading, searching, and editing of your personal vault.

  • Project
    bitbonsai/mcpvault

    A local MCP server that enables AI assistants to read, search, and edit notes within an Obsidian vault directly via the file system.

  • Project
    cyanheads/obsidian-mcp-server

    obsidian-mcp-server is an MCP‑compatible Node server that proxies the Obsidian Local REST API, exposing 14 tools (read, search, edit, tag management, command‑palette) and three resources over STDIO or HTTP. It lets LLM agents retrieve and modify vault notes safely, with optional path‑based read/write gating, delete confirmation, and optional command execution.

  • Project
    coddingtonbear/obsidian-local-rest-api

    Obsidian‑Local‑REST‑API is an Obsidian plugin that runs a secure HTTPS server exposing a RESTful and MCP‑based API. It lets external scripts or AI agents perform full CRUD on vault files, patch specific headings or front‑matter, search with fuzzy or JsonLogic queries, list/execute Obsidian commands, open notes in the UI, and more—all authenticated via a bearer token. The same capabilities are also reachable via the Model Context Protocol, enabling seamless integration with AI tools like Claude Code, Claude Desktop (via `mcp‑remote`), and Cursor.

  • Project
    StevenStavrakis/obsidian-mcp

    A local Model Context Protocol server that enables AI assistants to safely read and modify Obsidian vaults by interacting directly with Markdown files.