M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
M

Mindwtr MCP Server

Official
Let AI assistants read and manage your local Mindwtr GTD tasks.
How to install ↓ Compare GitHub source repository ↗
Category
Other
Stars
★ 2.4k Very popular
Transport
stdio (local process)
Runtime
Node.js
Credentials
Optional API key
License
AGPL-3.0
Last commit
57FMRS · C

This is the official MCP server bundled with the Mindwtr open-source GTD app (version 1.1.4, AGPL-3.0 license), distributed as the npm package mindwtr-mcp over stdio, enabling local task automation with optional read-only self-hosted Cloud access.

Strongest · Maintenance 16/20 Weakest · Reliability 8/20

Reliability
8/20
Security and permissions
10/20
Maintenance
16/20
Documentation
11/20
Setup experience
12/20
Why each score
Reliability 8/20
The manifest declares a stdio transport and an npm package ([email protected]) that can in principle be launched via npx, but the provided material contains no tool list, tool parameter schemas, or any test/CI content specific to the MCP server. The README's CI badge points to repo-wide CI and does not demonstrate coverage of the MCP init/tool-list handshake or error paths. Per the static calibration rule, without verifiable execution evidence (real CI plus committed tests covering the server's key paths) the score cannot exceed 12; given no evidence comparing declared vs. actual tool behavior, this is judged as a plausible happy path with untested edge cases and unverified tool behavior.
Security and permissions 10/20
Credentials are passed via environment variables (optional MINDWTR_MCP_CLOUD_TOKEN bearer token, optional MINDWTR_DB_PATH file path); no hardcoded secrets or plaintext token examples are present, and there is no evidence of known vulnerabilities or fabricated identity (no red-line issue). However, the manifest describes 'automation' of local data (implying write/execute capability) versus explicitly read-only Cloud access, and no tool list is provided to confirm whether destructive actions (delete, bulk operations, backup restore) require confirmation or are scoped to least privilege. Data boundaries and authorization scope are incompletely disclosed, placing this in a moderate-risk band rather than a red-line failure.
Maintenance 16/20
The project has a clear license (AGPL-3.0), explicit version iteration (1.1.4), a GitHub Actions CI badge, a contribution guide (CONTRIBUTING.md), a release notes index, a Discord community, and sponsorship channels, with 57 open issues and no archival — indicating ongoing maintenance and clear ownership (dongdongbh). No dedicated security-response policy file (e.g., SECURITY.md) was found in the evidence, but overall governance and update cadence are clear, warranting a relatively high score.
Documentation 11/20
The README documents the overall application thoroughly and provides install commands for the MCP server (npm install -g mindwtr-mcp / npx -y mindwtr-mcp) plus a link to an external docs site. The manifest gives brief descriptions for each environment variable. However, the supplied material lacks specific tool parameters, response formats, rate limits, cost information, or MCP-server-specific troubleshooting content — these largely rely on an external docs site whose content was not provided, leaving hidden assumptions and verification gaps.
Setup experience 12/20
Installation steps are simple and clear (global npm install or direct npx invocation), the stdio transport type is explicit, and environment variables are all optional with brief descriptions, giving a low barrier to entry. However, no concrete client configuration JSON example (e.g., for Claude Desktop) is given in the provided material, and there is no verifiable end-to-end connection evidence. Per the static calibration rule, in the absence of verifiable execution evidence this dimension is capped at 15, so it is judged usable but still requiring manual client-config assembly by the user.

Static review · not runListed 2026-08-06

Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkConnects to a database

Best for

  • Existing Mindwtr users who want AI assistance with task capture and processing
  • Developers or power users comfortable configuring a local stdio MCP server
  • Self-hosters running a Mindwtr Cloud instance who need read-only querying

Not for

  • Users without Mindwtr installed or without a local SQLite database
  • Teams needing multi-user collaborative writes via Cloud (Cloud mode is read-only)
  • Non-technical users unfamiliar with npm package installs and environment variables

Required permissions

  • Read/write access to the local Mindwtr SQLite database file (path set via MINDWTR_DB_PATH)
  • Network access to a self-hosted Mindwtr Cloud URL when Cloud mode is configured (read-only)
  • A bearer token credential for Cloud API access

Risks and side effects

  • The server has write access to the local database, so erroneous AI-issued changes could corrupt or pollute local task data
  • Storing the Cloud bearer token insecurely in environment variables risks credential exposure
  • Task and project data is exposed to whichever AI model is connected, with no additional sandboxing on scope

Setup

Before you start

Runtime：Node.js

MINDWTR_MCP_CLOUD_TOKEN optionalsecret Optional Bearer token for read-only access to a self-hosted Mindwtr Cloud instance; obtain it from your Cloud deployment.
Other optional settings (2)
MINDWTR_DB_PATH optional Optional path to the local Mindwtr SQLite database; omit to use the default location.
MINDWTR_MCP_CLOUD_URL optional Optional base URL of a self-hosted Mindwtr Cloud instance for read-only Cloud mode.
  1. Make sure Node.js and npm are installed.
  2. In your MCP client's stdio server configuration, add the command npx -y mindwtr-mcp (or run npm install -g mindwtr-mcp first and reference the mindwtr-mcp command directly).
  3. Optionally set MINDWTR_DB_PATH to point to your local Mindwtr SQLite database file if it isn't in the default location.
  4. For self-hosted Cloud read-only mode, optionally set MINDWTR_MCP_CLOUD_URL (your self-hosted Cloud base URL) and MINDWTR_MCP_CLOUD_TOKEN (bearer token).
  5. Save the config and restart the client to confirm the connection.
claude_desktop_config.json 
{"mcpServers": {"mindwtr": {"command": "npx", "args": ["-y", "mindwtr-mcp"]}}}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "mindwtr": {
      "command": "npx",
      "args": [
        "-y",
        "mindwtr-mcp"
      ]
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add mindwtr -- npx -y mindwtr-mcp

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

Configure {"mcpServers": {"mindwtr": {"command": "npx", "args": ["-y", "mindwtr-mcp"]}}} in your MCP client and start a session; Mindwtr's task tools should appear in the client's tool list, and asking it to list inbox tasks should return your local data.

Troubleshooting

  1. If tasks don't appear, verify MINDWTR_DB_PATH points to the correct local Mindwtr SQLite file
  2. Confirm npm/npx can reach the registry to install mindwtr-mcp
  3. For Cloud mode, check that MINDWTR_MCP_CLOUD_URL and MINDWTR_MCP_CLOUD_TOKEN are set correctly and the self-hosted Cloud instance is reachable
  4. Check the MCP client's stdio connection logs for startup or handshake errors

Things to try

Once connected, you can ask your AI assistant things like:

  • Add 'order birthday cake for Mom' to my Mindwtr inbox.
  • List my tasks scheduled for today and my next actions.
  • Turn 'plan the birthday party' into a project with small steps.
  • Search for all tasks marked as waiting for someone.

Use cases

Have an AI assistant help clarify the Inbox, add, or process GTD tasks during a chat
Ask the AI to surface overdue or waiting-for items to support a weekly review
Get a read-only task overview from a self-hosted Mindwtr Cloud instance

Supported clients

Claude Desktop

Listed from the project's documentation, not tested by this site.

Overview

Mindwtr MCP Server is the official first-party component bundled with the Mindwtr open-source GTD app, distributed as the npm package mindwtr-mcp running over stdio. It lets AI assistants read and manage tasks and projects in a local Mindwtr SQLite database, with optional read-only access to a self-hosted Mindwtr Cloud instance.

Similar servers

Telegram MCP Server 70 · B

Telegram MCP server powered by Telethon to let MCP clients read chats, manage groups, and send/modify messages, media, contacts, and settings.

★ 2.0k · Tools 76 Compare with this →
Compare with other MCP servers

Source revision c30388ad2244 Data synced 2026-11-04 Read the FMRS scoring method