What is the Continuous Development Cycle?

It is an autonomous workflow for Claude Code that combines planning, execution, and expansion. It creates a plan, writes code, commits changes to Git, and then proposes new features to keep the development loop running indefinitely.

Can I stop the autonomous loop once it starts?

Yes, you can stop the cycle at any time by using the /cancel-continuous-dev command, which removes the state file and allows for a normal exit while preserving all work and commits.

How does the auto-commit feature work?

The skill monitors your task_plan.md file. When all tasks in a phase are marked as complete and the status is updated, it automatically stages changes, creates a descriptive commit message, and pushes the code to GitHub.

What are the 3-file patterns used in this skill?

The skill utilizes task_plan.md for roadmap tracking, findings.md for technical research and documentation, and progress.md for logging session activities and test results.

Continuous Development Cycle

Name: Continuous Development Cycle
Author: inakam

byinakam

Productivity & Workflow

Automates the end-to-end software development lifecycle through autonomous planning, iterative execution, and perpetual feature expansion.

About

The Continuous Development Cycle skill transforms Claude into an autonomous software engineer capable of managing complex, multi-phase projects. By integrating a structured three-file planning system with the Ralph-loop execution pattern, it handles the entire workflow from initial research and task breakdown to code implementation and automated Git commits. Its standout feature is the 'Expansion' phase, where the agent analyzes completed work to propose and execute the next logical improvements, creating a self-sustaining development loop that continues until the project reaches maturity.

Key Features

State persistence across sessions to handle long-running complex tasks
Structured 3-file planning system (task_plan.md, findings.md, progress.md)
Automatic Git commits and GitHub repository synchronization on phase completion
AI-driven expansion proposer for perpetual development cycles
Autonomous Ralph-loop integration for iterative code execution
0 GitHub stars

Use Cases

Building full-stack applications from scratch with minimal human intervention
Continuous refactoring and optimization of existing codebases
Autonomous maintenance and feature expansion of greenfield projects

About

Key Features

State persistence across sessions to handle long-running complex tasks
Structured 3-file planning system (task_plan.md, findings.md, progress.md)
Automatic Git commits and GitHub repository synchronization on phase completion
AI-driven expansion proposer for perpetual development cycles
Autonomous Ralph-loop integration for iterative code execution
0 GitHub stars

Use Cases

Building full-stack applications from scratch with minimal human intervention
Continuous refactoring and optimization of existing codebases
Autonomous maintenance and feature expansion of greenfield projects