lc-dir

Instantly copy all relevant Python files from any folder in your project to your clipboard—perfect for use with AI code assistants and LLMs.

Why use lc-dir?

When working with LLMs (like ChatGPT, Claude, or Copilot), you often need to share only part of your codebase—a single directory, a feature, or a component, not the whole repo.

lc-dir lets you do exactly that.

• No more zipping folders or hand-pasting dozens of files.
• No more sending the entire project when only one area matters.
• Get context for refactoring, debugging, or reviewing code fast.

Key Features

• Easy: Instantly copies all .py files under any directory (recursively) to your clipboard, ready for LLMs.
• Flexible: Works from any subfolder, or specify any target directory.
• Search: Give it a folder name; it’ll find the match for you.
• Integrated: Uses your llm-context workflow for formatting and clipboard.

Installation

first install llm-context CLI if you haven't already:

pipx install llm-context

Then, install lc-dir:

pipx install lc-dir

(or use pip install --user lc-dir if you prefer)

Requirements:

• Python 3.10+
• The llm-context CLI, initialized in your project (run lc-init once per repo).

How to Use

Example: Copy all .py files from the current directory and below:

lc-dir

Example: Copy all .py files from a specific subdirectory (recursively):

lc-dir path/to/subdir

Example: Search for a folder by name, even if you’re not sure where it is:

lc-dir common

If more than one match is found, you’ll be prompted to choose.

Typical Workflow

1. Navigate anywhere in your project directory tree.

2. Run lc-dir [optional-subfolder]

3. Paste in your LLM interface.

   • The clipboard will contain all relevant code, formatted for LLMs, just as if you ran llm-context but limited to your chosen scope.

Use Cases

• LLM-powered code review: Share just the area you want help with.
• Refactoring: Provide only the module or feature you’re working on.
• Bug fixing: Send just the files around the problem.
• Onboarding: Give teammates or LLMs a focused view of part of the repo.

Troubleshooting

• If the clipboard isn’t updated, ensure you’ve initialized your repo with lc-init.
• If you see “ModuleNotFoundError”, re-install with pipx install lc-dir and check your Python version.
• You must have llm-context CLI installed in the project for this to work.

License

MIT

Contributions and issues welcome!