Will I lose my current work by using this skill?

No, the script specifically clones the later half of the conversation, ensuring that your most recent messages, instructions, and code changes are preserved.

Does this skill delete my original conversation history?

No, it creates a new session entry in your history file, leaving the original, full-length conversation completely untouched and accessible.

What is the primary benefit of the Half-Clone skill?

It significantly reduces the number of tokens sent to the AI model, which lowers costs and increases response speed while keeping your most recent work context intact.

How do I access a conversation after it has been half-cloned?

You can access the new session by running the 'claude -r' command and looking for the entry marked with the [HALF-CLONE] tag and a timestamp.

How do I ensure I am cloning the right conversation?

The skill includes a mandatory preview step that allows you to verify the session ID and check that the first and last messages match your current work before proceeding.

Conversation Half-Clone

Name: Conversation Half-Clone
Author: ykdojo

byykdojo

•

1,975

Productivity & Workflow

Reduces token usage and improves performance by cloning only the most recent half of your Claude Code conversation history.

About

The Half-Clone skill optimizes your Claude Code sessions by discarding older, irrelevant context while preserving your most recent work and decisions. This utility is essential for developers working on long-running tasks who want to minimize token costs, avoid hitting context window limits, and maintain high model performance. By intelligently pruning the conversation history, it provides a safe way to continue development with a leaner context, ensuring the AI stays focused on current objectives without being bogged down by early-session noise.

Key Features

1,975 GitHub stars
Preservation of recent work state and immediate project context
Seamless integration with Claude Code session history and project paths
Automatic token reduction by pruning older conversation history
Safety preview mode to verify session data before cloning
Automatic session tagging for easy retrieval via the resume command

Use Cases

Optimizing long-running coding sessions that have become slow or expensive due to large context
Forking a conversation to try a different architectural approach from the current state
Managing model context limits during massive refactoring projects without losing recent progress

About

Key Features

1,975 GitHub stars
Preservation of recent work state and immediate project context
Seamless integration with Claude Code session history and project paths
Automatic token reduction by pruning older conversation history
Safety preview mode to verify session data before cloning
Automatic session tagging for easy retrieval via the resume command

Use Cases

Optimizing long-running coding sessions that have become slow or expensive due to large context
Forking a conversation to try a different architectural approach from the current state
Managing model context limits during massive refactoring projects without losing recent progress