mission-control  by MeisnerDan

Open-source task management orchestrating AI agents for solo entrepreneurs

Created 2 weeks ago

New!

263 stars

Top 96.9% on SourcePulse

GitHubView on GitHub
Project Summary

Mission Control is an open-source task management system designed for the "agentic era," specifically targeting solo entrepreneurs who delegate tasks to AI agents. It addresses the chaos of managing multiple AI agents across various projects by providing a structured command center with a visual dashboard, agent inboxes, and reporting protocols, enabling users to delegate work effectively and maintain oversight without micromanagement.

How It Works

Mission Control employs a local-first, agent-first architecture, storing all project data—tasks, goals, agents, and messages—in plain JSON files. This eliminates database dependencies and cloud reliance, ensuring data privacy and control. AI agents interact directly with these files via a token-optimized API, reading assignments and writing progress reports or seeking decisions. This approach allows any file-aware AI coding tool, such as Claude Code, Cursor, or Windsurf, to integrate seamlessly, with agents executing tasks through local CLI commands.

Quick Start & Requirements

  • Prerequisites: Node.js v20+, pnpm v9+. Claude Code is recommended for agent execution (npm install -g @anthropic-ai/claude-code).
  • Install: Clone the repository, navigate to the mission-control directory, and run pnpm install.
  • Run: Execute pnpm dev to start the web UI. Access http://localhost:3000 and load demo data for an initial experience.
  • Docs: Repository Link

Highlighted Details

  • Local-First Architecture: All data resides in local JSON files, ensuring no cloud dependency or API keys are required.
  • Agent-Centric Design: Features a token-optimized API for agents, enabling efficient communication and task execution.
  • Task Management Suite: Includes an Eisenhower Matrix for prioritization, Kanban boards, Goal Hierarchy tracking, and a Brain Dump feature.
  • Automated Agent Operations: Supports an Autonomous Daemon for background task polling, One-Click Execution for tasks, Continuous Missions for project workflows, and Session Resilience with loop detection.
  • Cost & Usage Tracking: Monitors and displays token usage and associated costs for every Claude Code session.
  • Extensible Agent Crew: Comes with 5 built-in agents and allows for unlimited custom agent creation with a Skills Library for reusable knowledge modules.

Maintenance & Community

This project is described as a personal endeavor built to organize the author's work. While it welcomes contributions, there are no explicit mentions of dedicated community channels (like Discord/Slack), sponsorships, or a formal roadmap beyond features listed as planned in the README.

Licensing & Compatibility

Mission Control is released under the MIT License, permitting unrestricted use, modification, and distribution, including for commercial purposes. It runs locally and integrates with AI tools via the filesystem and CLI.

Limitations & Caveats

The project is provided "as-is" with no warranties, and users assume all risk. Full agent execution capabilities rely on having Claude Code installed and configured locally. Features like cloud sync, Docker support, and a PWA version are listed on the roadmap but are not yet implemented.

Health Check
Last Commit

4 days ago

Responsiveness

Inactive

Pull Requests (30d)
6
Issues (30d)
12
Star History
267 stars in the last 18 days

Explore Similar Projects

Feedback? Help us improve.