monoco-toolkit 0.3.5

Agent Native Toolkit for Monoco - Task Management & Kanban for AI Agents

Monoco Toolkit

The Operating System for Agentic Engineering.

Stop chatting. Start building. Monoco grounds your AI Agents into deterministic, validatable, and shippable engineering workflows.

---

⚡️ The Gap: "Chat" is not Engineering.

Generating code is easy. Engineering—managing dependencies, maintaining state, validation, and version control—is hard.

Most Agent workflows today are fragile "chats". Monoco turns them into Software Engineering Processes. It acts as the "BizOps" layer for your Agents, ensuring that every line of code generated is tracked, reviewed, and aligned with the project roadmap.

🌟 Core Philosophy

1. Co-pilot by Design

Just-enough Automation. Monoco abstracts away the tedious details (Git ops, state tracking) while keeping you securely in the driver's seat. Agents can suggest and build, but you always have the final say on what gets merged.

2. Battle-Tested Best Practices

Senior Engineer Intuition. Monoco enforces Issue Driven Development (IDD) and standard Git Workflows. We bake the rigorous habits of effective software teams into the Agent's core loop, ensuring every line of code is traceable and reviewed.

3. Radical Transparency (Dogfooding)

Process as Product. Monoco is built by Monoco. We believe in open-sourcing not just the code, but the engineering process itself. Every design decision, interaction log, and failure is public—providing a live blueprint for Agentic Engineering.

🚀 Quick Start

1. Installation

Monoco is available as a Python CLI tool.

pip install monoco-toolkit

2. Workspace Initialization

Turn any directory into a Monoco Workspace. This creates the .monoco config and the Issues/ directory.

monoco init

3. Agent Synchronization

Crucial Step: This injects the "Monoco System Neural Network" (System Prompts & Skills) into your agent configuration files (e.g., GEMINI.md, CLAUDE.md).

monoco sync

4. The Engineering Loop (Agent-First)

In Monoco, you don't need to memorize CLI commands. The Agent is your DevOps Engineer.

1. Chat: Tell your Agent in the chatbox (e.g., "Implement Dark Mode").
2. Plan: The Agent investigates and proposes an Issue Ticket for your review.
3. Build: Once approved, the Agent creates a branch, writes code, and submits changes.
4. Ship: You accept the results. The Agent handles the merge and closure.

---

📦 Extension for VS Code

The Monoco VS Code Extension is the primary visual interface for the toolkit.

• Install from Marketplace: Search for Monoco.
• Keybinding: Cmd+Shift+P -> Monoco: Open Kanban Board.

🛠️ Tech Stack & Architecture

• Core: Python (CLI & Logic Layer)
• Extension: TypeScript (VS Code Client & LSP)
• Data: Local Filesystem (Markdown/YAML)

🤝 Contributing

Monoco is designed for the community. We welcome contributions to both the core CLI and the VS Code extension.

📄 License

MIT © IndenScale