obsidian-vault

This MCP server grants AI assistants read and write access to your Obsidian vault, with specialized support for PARA organization, Obsidian Tasks, and daily journaling workflows. It exposes a suite of 16 tools organized into Read, Analytics, Write, and Daily Journal categories, enabling you to read notes by path or title, search content with full-text queries, resolve wikilinks and backlinks, extract task statistics, gather topic information, and create notes, inbox captures, daily entries, or attachments within your vault. Use cases include building AI-assisted vault navigation, topic clustering across PARA areas, automated task reporting, and smart note creation that respects PARA locations and folder structures. The tools return structured metadata and content suitable for downstream AI reasoning and document synthesis, including summaries, snippets, and per-note statistics.

To use the tools, connect your MCP-enabled AI assistant to the Obsidian Vault MCP server and call the desired tool by name with the required parameters. For example, invoke obsidian_search_notes with a query to locate relevant notes, or obsidian_get_task_stats to analyze tasks within a specific folder. The server understands PARA-based locations (inbox, projects, areas, resources, archive) and supports optional filters like tag matching, date ranges, and content snippets.

Typical workflows include: (1) Topic gathering: use obsidian_gather_topic to collect and group notes related to a topic by PARA location, (2) Daily journaling: use obsidian_create_daily_note to log daily insights, (3) Quick capture: use obsidian_create_inbox_note to funnel ideas into INBOX for later processing, and (4) Project activity: run obsidian_get_project_activity to surface stale projects with no recent activity.

Prerequisites:

• Node.js (recommended v14+ or LTS) and npm installed on your machine
• Internet access to install the MCP package via npx

Installation steps:

1. Verify Node.js and npm are available

   • node -v
   • npm -v

2. Install or run the Obsidian Vault MCP server via npx (no global install required)

   • npx -y obsidian-vault-mcp

3. Configure environment if needed (optional)

   • You can set environment variables to customize behavior, such as logging level or vault path, depending on the MCP package capabilities. Example placeholders:
     • OBSIDIAN_VAULT_PATH=/path/to/vault
     • MCP_LOG_LEVEL=info

4. Run the server and connect your MCP client

   • The command above will start the MCP server. Follow the package’s prompts to expose the API endpoints as defined by the MCP spec.

5. Optional: integrate with your AI assistant or orchestration layer following your platform’s MCP client docs.

Tips and common issues:

• Ensure your Obsidian vault is accessible by the MCP server process (correct permissions and vault path).
• If you encounter missing tool definitions, verify you’re using the latest obsidian-vault-mcp package and that your MCP client is targeting the correct server name (obsidian-vault).
• When using date filters, ISO date formats (YYYY-MM-DD) are expected for created/modified fields.
• For write actions, ensure your assistant has proper prompts to provide required parameters (e.g., title, para_location, content) to avoid failed calls.
• If running behind a firewall or proxy, ensure the MCP server can communicate with your AI client over the configured port.
• Environment variables can customize vault path, logging, and feature toggles if supported by the package; check the package docs for available options.