MeerTech product · Productivity · Applied AI

ClearMind: productivity workspace with an MCP server.

Tasks, projects, notes, habits and goals in one workspace that works offline, with an AI assistant and an MCP server so agents such as Claude can act on your data through scoped OAuth access.

Deployed Checked October 2026
Three ClearMind Android screens: a project overview, today's task list and linked notes
ClearMind · Web · Android · API · MCP
01What it does

A productivity workspace that AI agents can work in safely.

Status
Deployed. A client, institution or the public is using it in production today.
Built for
Individuals, developers and professionals
Platforms
Web, Android, API, MCP

Key capabilities

  1. 01Tasks with natural-language quick add, priorities, recurrence and saved filters
  2. 02Projects with phases, risks, KPIs and health status
  3. 03Offline-first sync shared by web, Android, home-screen widgets, API and CLI
  4. 04MCP server and REST API behind OAuth with scoped tokens
  5. 05Notes, habits, calendar and a daily time-block planner
  6. 06Iris, an AI planning assistant built on Google Gemini
02Problem and approach

The operational challenge

Planning is scattered, and assistants have no safe way in.

Personal productivity tools split tasks, notes and planning across separate apps, and most give AI assistants no controlled way to help. People either paste context into a chat by hand or give an assistant access to far more than it needs.

Who uses it. Individuals, developers and professionals who plan their work in one place and increasingly hand parts of it to AI assistants.

The engineering approach

One workspace, open to agents on your terms.

ClearMind brings tasks, projects, notes, habits, calendar and goals into one workspace that keeps working offline and syncs across the web, Android and home-screen widgets. The same data is available through a REST API, a command-line tool and a Model Context Protocol server, each behind OAuth with scoped tokens, so an assistant can read or update tasks without full access to the account.

03Architecture

How it is put together.

  • React 19
  • TypeScript
  • Vite
  • Firebase
  • Expo
  • Model Context Protocol
  • Gemini API
  • Vercel
  1. 01Clients Installable web app, Android app with five home-screen widgets, and a CLI
  2. 02Sync Offline first, with per-field clocks, an outbox and delta listeners, so concurrent edits merge instead of overwriting each other
  3. 03Agent access MCP server and REST API with OAuth discovery and scoped tokens
  4. 04AI The Iris assistant runs on the Gemini API and helps with planning; it is not used as a source of record
  5. 05Hosting Vercel for the web app and API, Firebase for data
Fig · ClearMind architectureSchematic. Components as built; no measured values.

Where it stands

Live on the web at clearmind.one and on Google Play for Android. The iOS app has been built and submitted to Apple but is not yet on the App Store. A macOS desktop app is in development.

Contact

Talk to the team behind ClearMind.

Describe the infrastructure, the workflow that costs you most, and your connectivity as it really is. The first call is with an engineer.