Features

Everything taskpapr can do — from the core board to integrations and automation.


taskpapr is a minimal, paper-inspired task board that runs locally or on a server you control. It has no accounts to create, no cloud lock-in, and no noise. Below is a map of everything it can do.

Feature pages

Page What it covers
The Board Infinite canvas, tiles, tasks, WIP state, search, keyboard shortcuts
Recurring Tasks Due dates, spinning plates urgency heat, dormancy, task rot, snooze
Goals Outcome tracking, goal smart-tiles, assigning tasks to goals
Telegram Daily digest, self-service connect flow, quick-capture from chat
API & Webhooks REST API reference, webhook actions, n8n automation examples
MCP Server AI assistant integration — Claude Desktop, Cline, 9 available tools
Import & Export Backup, restore, merge vs replace, curl examples

At a glance

The board is an infinite canvas. Tiles (columns) sit wherever you put them. Tasks live inside tiles. Nothing is forced into a grid.

Recurring tasks have a visual urgency system — amber → orange → red as a due date approaches. One-off tasks have a separate rot system — they quietly age to parchment if left untouched. Both are purely visual; no notifications, no badges.

Goals are outcomes, not projects. Assign tasks to a goal and see them collected in a smart-tile on the canvas. Track progress by task count.

Telegram sends a daily digest of tasks due today and tomorrow. You can also send plain text to the bot to create tasks instantly from your phone.

The API is a standard REST API with Bearer token auth. The webhook endpoint accepts four actions (add_task, complete, mark_wip, delete_task) and works with n8n, Zapier, Make, or any HTTP client.

The MCP server exposes your board as tools for Claude Desktop, Cline, or any MCP-compatible AI assistant. Ask it what’s on your board, add tasks, mark things done — in plain language.

Export/import produces a portable JSON file. Import in merge or replace mode. Use it for backups, migrating between instances, or loading demo data.


Want to suggest a feature? Open an issue on GitHub.


Table of contents