mcp-obsidian-via-rest
MCP Server connected to Obsidian Vault with Local REST API. (re-write it completely with BUN to make it faster and more lightweight)
claude mcp add --transport stdio oleksandrkucherenko-mcp-obsidian-via-rest docker run --name mcp-obsidian --rm -i -p 3000:3000 -e API_KEY -e API_URLS -e DEBUG ghcr.io/oleksandrkucherenko/obsidian-mcp:latest \ --env DEBUG="mcp:*" \ --env API_KEY="<secret_key>" \ --env API_URLS="["https://127.0.0.1:27124","https://172.26.32.1:27124","https://host.docker.internal:27124"]"
How to use
This MCP server exposes an Obsidian REST API integration to AI assistants via the MCP framework. It provides tools to retrieve note contents and to search Obsidian vaults, including a semantic search capability. The available tools are: get_note_content, which fetches the content and metadata of a given Obsidian note; obsidian_search, which performs a text-based search across notes using a query; and obsidian_semantic_search, which performs a semantic search over notes using a query. To use the server, deploy it (for example via Docker as shown in the configuration), ensure your API_KEY and API_URLS are configured, and then interact with the tools through your MCP client or prompt-driven AI workflow. The multi-URL configuration enables automatic failover and self-healing, selecting the fastest available Obsidian API URL and reconnecting on failure, with transport options including stdio for local use and HTTP JSON-RPC for remote access.
How to install
Prerequisites:
- Docker installed and running (or adapt to your preferred environment listed in mcp_config).
- Access to an Obsidian vault via Obsidian REST API (API endpoint(s) reachable from the MCP server).
Installation steps:
- Pull and run the MCP Obsidian Docker image (as in the example):
# Create a suitable environment file or set env vars directly
export API_KEY=<your_api_key_here>
export API_URLS=["https://127.0.0.1:27124","https://host.docker.internal:27124"]
export DEBUG=mcp:*
# Run via Docker (as shown in the config snippet)
docker run --name mcp-obsidian --rm -i \
-p 3000:3000 \
-e API_KEY \
-e API_URLS \
-e DEBUG \
ghcr.io/oleksandrkucherenko/obsidian-mcp:latest
- If you prefer a different transport, adjust the mcpServers entry to use your chosen method (stdio or http) per the Available transports in the Configure MCP section of the README.
- Verify the MCP server is up by querying the health or invoking one of the Tools endpoints (get_note_content, obsidian_search, obsidian_semantic_search) via your MCP client.
Prerequisites recap:
- Docker or compatible container runtime
- Access to Obsidian REST API endpoints (API_URLS)
- A valid API_KEY if required by your Obsidian REST setup
Additional notes
Tips and notes:
- Use API_URLS with a JSON array of endpoints for automatic failover; the MCP server will test and select the fastest URL on startup.
- The DEBUG environment variable can help troubleshoot by enabling mcp:* logs.
- Ensure port 3000 (or your configured port) is accessible if you expose the HTTP transport.
- If you rotate API keys, update API_KEY accordingly and restart the container to apply changes.
- For local development, stdio transport may be the simplest option, while HTTP transport is better for remote clients.
- When using multi-URL failover, the server will monitor health every 30 seconds and switch to a healthy URL if the current one becomes unavailable.
Related MCP Servers
MiniMax -JS
Official MiniMax Model Context Protocol (MCP) JavaScript implementation that provides seamless integration with MiniMax's powerful AI capabilities including image generation, video generation, text-to-speech, and voice cloning APIs.
mcp-obsidian
MCP server for Obsidian (TypeScript + Bun)
akyn-sdk
Turn any data source into an MCP server in 5 minutes. Build AI-agents-ready knowledge bases.
mcp-obsidian-advanced
Advanced MCP server for interacting with Obsidian via the Local REST API community plugin. It empowers AI agents (like Claude) to deeply understand your vault's structure, links, and content—beyond basic read/write operations.
project-thoth
An advanced AI-powered research assistant system built for flexibility, extensibility, and user control. Thoth combines intelligent automation with comprehensive customization — every prompt, schema, and source can be tailored to your workflow.
obsidian-http
First HTTP-native MCP server for Obsidian - Compatible with any MCP client (Claude Code, Codex, Gemini, etc.) without stdio bugs