Skip to content
Conduit

mcp server

sharpei

Sharpei Todo Management

bnaylor/sharpeicommunitylistedunreviewed

Set up

Runs on your computer. Pick your agent and operating system.

This project does not publish an install method we can turn into a config. Follow the README instead.

README

synced Oct 26, 2026

Sharpei

A focused, local-first TODO application inspired by MyTinyTodo.

Sharpei runs entirely on your machine with no cloud dependencies, stores data in SQLite for easy AI/agent access, and provides a clean, responsive UI for managing tasks, and an MCP server for direct AI agent manipulation.

Features

  • Local-first - Runs on localhost, no internet required
  • SQLite storage - Simple, portable database that AI tools can read
  • Calendar View - Interactive monthly grid to visualize deadlines
  • Bulk Actions - Multi-select tasks for group updates (move, priority, delete)
  • Task Recurrence - Automatically repeating tasks (daily, weekly, custom)
  • Smart Categories - Save search queries as persistent sidebar items
  • Advanced Search - Filter by status (is:overdue), priority, and more
  • PWA Support - Install as a standalone desktop app with offline caching
  • Data Portability - JSON export/import and automated daily backups
  • Markdown notes - Capture details about your tasks in markdown syntax
  • Subtasks - Break down tasks into smaller items
  • Hashtags - Tag tasks for easy filtering
  • Quick-add syntax - Rapidly create tasks with !priority #tags @dates *recur >Category
  • MCP server - AI assistant integration via Model Context Protocol
  • Dark Mode - Full support for light and dark themes

Quick Start

# Clone the repo
git clone https://github.com/bnaylor/sharpei.git
cd sharpei

# Create virtual environment
python3 -m venv venv
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Run the app
python sharpei.py

The app opens automatically at http://127.0.0.1:8000

Remote Hosting

Sharpei is designed for local-first use, but you can also host the backend on a remote server and connect to it from any browser or PWA. Local-first remains the default.

Running the Backend for Remote Access

To allow remote connections, start the backend with the --host and --api-key arguments:

python sharpei.py --host 0.0.0.0 --api-key YOUR_SECRET_KEY
ArgumentDescriptionDefault
--hostInterface to bind to (0.0.0.0 for all)127.0.0.1
--portPort number to listen on8000
--api-keySecret key required for API accessNone
--ssl-certPath to SSL certificate (for HTTPS)None
--ssl-keyPath to SSL key (for HTTPS)None
--no-browserSkip opening the browser on startupfalse

Connecting from the UI

  1. Open the Sharpei UI.
  2. Click the Settings (gear icon) in the top right corner.
  3. Under Remote Backend:
    • Enter your Remote API URL (e.g., https://sharpei.example.com:8000).
    • Enter your API Key.
  4. Click Save Settings.

The UI will now use the remote backend for all operations. This configuration is saved in your browser's local storage.

Quick-Add Syntax

Create tasks rapidly from the input field:

Call dentist @tomorrow !high #health
Submit timesheet *weekly @friday >Work
Water plants *3d @tomorrow
ElementSyntaxExamples
Priority!high !low!high, !h, !low, !l
Hashtags#tag#work #urgent
Due date@date@today, @monday, @+3d, @2025-02-15
Recurrence*pattern*daily, *weekly, *monthly, *3d, *2w
Category>name>Work, >Personal

See doc/quick-add.md for full documentation.

Keyboard Shortcuts

ShortcutAction
Ctrl+Shift+AArchive completed tasks
Ctrl+Shift+HToggle archived tasks visibility
Ctrl+Shift+DToggle details view

MCP Server

Sharpei includes an MCP server for AI assistant integration:

python mcp_server.py

See doc/mcp.md for full documentation.

Testing

# Run all tests (125 total)
pytest

Documentation

Credits

Built with assistance from Gemini 3 and Claude Opus 4.5.

License

Apache 2.0 License - see LICENSE for details.

↑↓ navigate openesc close