vibefree 0.1.0

A tiny CLI initializer for vibe coding projects that generates .memory and .harness scaffolding.

vibefree

vibefree is a tiny Python CLI initializer for vibe coding projects. It creates a lightweight .memory and .harness scaffold so coding agents have a consistent place to read project context, rules, workflow notes, and quality expectations.

The 0.1.0 release focuses on one reliable command: vibefree init.

Features

• Creates .memory files for durable project knowledge.
• Creates .harness files for coding-agent instructions.
• Supports initializing the current directory or a named project directory.
• Fills missing files without overwriting existing files.
• Keeps the generated structure small, readable, and tool-agnostic.

Installation

Install from PyPI:

pip install vibefree

Usage

Initialize the current directory:

vibefree init

Initialize a named project directory:

vibefree init my-project

If my-project does not exist, it is created. If it already exists, it is reused.

Existing generated files are never overwritten. Missing files are added on later runs.

Generated Structure

.memory/
├── memory.md
├── rule.md
├── project.md
└── logs/
    └── .gitkeep

.harness/
├── AGENTS.md
├── ARCHITECTURE.md
├── WORKFLOW.md
├── QUALITY.md
└── plans/
    ├── active/
    │   └── .gitkeep
    └── completed/
        └── .gitkeep

Generated Files

• .memory/memory.md: Long-term project memory for durable facts, decisions, and constraints.
• .memory/rule.md: Rules for when agents should read or update memory.
• .memory/project.md: Basic project information, including the project name.
• .memory/logs/.gitkeep: Keeps the logs directory in version control.
• .harness/AGENTS.md: Agent entry file with the recommended reading order.
• .harness/ARCHITECTURE.md: Architecture notes and design decisions.
• .harness/WORKFLOW.md: Development workflow guidance.
• .harness/QUALITY.md: Quality, testing, and safety rules.
• .harness/plans/active/.gitkeep: Keeps the active plans directory in version control.
• .harness/plans/completed/.gitkeep: Keeps the completed plans directory in version control.

Development

Create a virtual environment:

python -m venv venv

Activate it on Windows PowerShell:

.\venv\Scripts\Activate.ps1

Activate it on macOS or Linux:

source venv/bin/activate

Install locally:

pip install -e .

Install development dependencies and run tests:

pip install -e ".[dev]"
pytest

Build distributions locally:

python -m build

Design Principles

• simple: keep the CLI and generated files easy to understand.
• agent-friendly: provide clear context files for coding agents and opencode-style workflows.
• no overwrite by default: preserve existing files and only create missing files.
• minimal first: start with the smallest useful structure and expand deliberately.

License

Apache License 2.0. See LICENSE.