Skip to content

CLI tool to copy many files, in order to provide better context for AI assistants. Helps to iterate faster while leveraging LLM assistants.

Notifications You must be signed in to change notification settings

markwitt1/concopy

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

concopy

concopy (concatenate-copy) is a compact Command Line Interface (CLI) tool designed to facilitate the seamless pasting of your codebase into your preferred Generative AI assistant. Whether you're using ChatGPT, Bard, or any company-internal Language Model, concopy enables you to write code efficiently by allowing you to select the specific context about your codebase to supply and iterate on the code using the AI assistants output.

demo

concopy proves particularly useful when you need to generate code for a specific segment of your codebase without sharing the entire codebase with the AI. It can sometimes outperform GitHub Copilot by enabling you to select the context of the code you wish to generate. Copilot's RAG occasionally focuses on irrelevant parts of the codebase, leading to undesired results.

copilot wrong result

As illustrated in the screenshot, Copilot only injects lines 1 to 32 into the prompt, despite the file containing many more lines.

Installation

Using Homebrew

You can install concopy using Homebrew by tapping into our repository and then installing the formula:

brew tap markwitt1/tap https://github.com/markwitt1/homebrew-tap
brew install concopy

Other Methods

Download the binary compatible with your platform and place it somewhere in your PATH.

Run chmod +x path/to/binary to make the binary executable.

Usage

By Command Line Arguments

Run concopy <file1> <file2> <folder1> to copy the content of the specified files to the clipboard.

Using .concopyuse

Add a .concopyuse file at your project root and populate it with the file/directory paths separated by line breaks.

For example:

go.mod
.github
main.go

NOTE: On the first run on macOS, the system may block the execution. You will need to navigate to System Settings -> Privacy & Security -> Security and allow the execution of the binary.

About

CLI tool to copy many files, in order to provide better context for AI assistants. Helps to iterate faster while leveraging LLM assistants.

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages