About
A connector that lets MCP clients, such as Claude Desktop, index and query any directory of Markdown notes—primarily Obsidian vaults—for seamless context retrieval.
Capabilities

The Obsidian Model Context Protocol server bridges the gap between an AI assistant and a personal knowledge base stored in Markdown. By exposing a vault directory to the MCP ecosystem, it lets Claude Desktop—or any compliant client—search, retrieve, and even modify notes on demand. This eliminates the need for manual copy‑and‑paste or API wrappers, making the vault a first‑class data source that the assistant can query as naturally as it queries a database.
At its core, the server translates standard file‑system operations into MCP resources. When a user asks the assistant to find information on a topic, the server performs a full‑text search across all Markdown files in the specified vault. Results are returned as structured objects containing file paths, snippets, and metadata such as tags or headings. This tight integration means developers can build workflows where the assistant not only answers questions but also suggests relevant notes, creates new entries, or updates existing ones—all without leaving the chat interface.
Key capabilities include:
- Full‑text search across an entire vault, with optional filtering by tags or folder structure.
- Contextual snippet extraction that preserves formatting, making it easy for the assistant to present information in a readable form.
- Metadata access (e.g., YAML front matter, tags) enabling advanced queries and dynamic content generation.
- Write‑back support for creating or updating notes, allowing the assistant to act as a collaborative writer.
- Custom input prompts for specifying vault paths, which can be shared across projects via VS Code settings.
Real‑world scenarios that benefit from this server are plentiful. Knowledge workers can ask their assistant to pull the latest research on a project, automatically generate meeting notes that link back to related vault pages, or even draft a blog post by aggregating insights from multiple notes. Educators can use it to pull lesson plans or student work, while developers might integrate the server into CI pipelines to auto‑document changes directly from Markdown sources.
Integration is straightforward: once installed, the MCP client discovers a new “Obsidian” tool. The assistant can then invoke it by name, passing search queries or file paths as arguments. Because the server follows MCP conventions, developers can compose complex chains—search → summarize → update—without writing custom connectors. This composability, coupled with the natural language interface of Claude, gives teams a powerful way to make their Markdown vaults conversational and actionable.
Related Servers
MarkItDown MCP Server
Convert documents to Markdown for LLMs quickly and accurately
Context7 MCP
Real‑time, version‑specific code docs for LLMs
Playwright MCP
Browser automation via structured accessibility trees
BlenderMCP
Claude AI meets Blender for instant 3D creation
Pydantic AI
Build GenAI agents with Pydantic validation and observability
Chrome DevTools MCP
AI-powered Chrome automation and debugging
Weekly Views
Server Health
Information
Explore More Servers
Weibo MCP Server
Real‑time Weibo data for AI applications
Peekaboo MCP Server
Fast macOS screenshots and AI-powered GUI automation
MCP Desk
Desktop client for Multi-Capability Platform servers
Buildkite MCP Server
Expose Buildkite pipelines to AI tools and editors
Mentor Mcp Server
MCP Server: Mentor Mcp Server
MCP Client
TypeScript SDK for JSON‑RPC MCP services