Obsidian Tasks MCP Server

Overview

The Obsidian Tasks MCP Server bridges the gap between a user’s Obsidian vault and AI assistants that speak the Model Context Protocol. It parses Markdown files in a vault, extracts tasks in the same format used by the popular Obsidian Tasks plugin, and presents them as structured JSON objects. This allows Claude or any MCP‑compatible assistant to query, filter, and manipulate a user’s task list as if it were an internal knowledge base.

The server solves the problem of “task‑parsing friction.” In a typical workflow, developers or knowledge workers need to pull tasks from scattered notes into dashboards, calendars, or AI‑driven planning tools. Without a dedicated interface, this requires manual copy‑paste or custom scripts that are brittle and hard to maintain. By exposing a clean, protocol‑based API, the MCP server lets AI assistants retrieve tasks on demand, apply complex filters, and even integrate with calendar or reminder services—all without the user leaving their note‑taking environment.

Key features include:

• Recursive extraction – scans a directory tree, returning every task with its unique identifier (file path plus line number) and metadata such as status, dates, tags, priority, and recurrence rules.
• Advanced querying – accepts Obsidian Tasks query syntax, enabling multi‑criteria searches (status, dates, tags, paths, descriptions, priority). This mirrors the native query language of the Obsidian plugin but exposes it over MCP.
• Rich metadata – Every task object contains fields for due dates, scheduled dates, start dates, creation timestamps, tags, priority levels, and recurrence rules, giving AI assistants full context to reason about deadlines or dependencies.
• Developer‑friendly JSON – Results are returned as plain JSON arrays, making it trivial to consume in any programming language or AI prompt.

Typical use cases span personal productivity and enterprise workflow automation. A developer can ask an assistant to “list all incomplete work tasks due next week” and receive a ready‑to‑use dataset that can be fed into a calendar sync or displayed in a dashboard. A project manager might query for high‑priority tasks with no due date and then ask the assistant to suggest deadlines. In an automated CI/CD pipeline, tasks extracted from documentation can trigger alerts or status updates in a team chat.

Integration with AI workflows is seamless: the MCP server can be launched as a local process or exposed via a network port, and any Claude‑compatible assistant can invoke the or tools. The assistant’s prompt can include natural language instructions, and the server will return structured data that the model can interpret or transform further. This tight coupling eliminates manual data wrangling and empowers AI assistants to act as true task‑management companions within the Obsidian ecosystem.