/pack-local

Pack a local codebase using Repomix with AI-optimized format.

When the user asks to pack a local codebase, analyze their request and run the appropriate repomix command.

User Intent Examples

The user might ask in various ways:

• "Pack this codebase"
• "Pack the src directory"
• "Pack this project as markdown"
• "Pack TypeScript files only"
• "Pack with compression and copy to clipboard"
• "Pack this project including git history"

Your Responsibilities

1. Understand the user's intent from natural language
2. Determine the appropriate options:
   • Which directory to pack (default: current directory)
   • Output format: xml (default), markdown, json, or plain
   • Whether to compress (for large codebases)
   • File patterns to include/ignore
   • Additional features (copy to clipboard, include git diffs/logs)
3. Execute the command with: npx repomix@latest [directory] [options]

Available Options

• --style <format>: Output format (xml, markdown, json, plain)
• --include <patterns>: Include only matching patterns (e.g., "src//*.ts,/*.md")
• --ignore <patterns>: Additional ignore patterns
• --compress: Enable Tree-sitter compression (~70% token reduction)
• --output <path>: Custom output path
• --copy: Copy output to clipboard
• --include-diffs: Include git diff output
• --include-logs: Include git commit history

Command Examples

Based on user intent, you might run:

# "Pack this codebase"
npx repomix@latest

# "Pack the src directory"
npx repomix@latest src/

# "Pack as markdown with compression"
npx repomix@latest --style markdown --compress

# "Pack only TypeScript and Markdown files"
npx repomix@latest --include "src/**/*.ts,**/*.md"

# "Pack and copy to clipboard"
npx repomix@latest --copy

# "Pack with git history"
npx repomix@latest --include-diffs --include-logs

Analyzing the Output

IMPORTANT: The generated output file can be very large and consume significant context.

If the user wants to analyze or explore the generated output:

• DO NOT read the entire output file directly
• USE an appropriate sub-agent to analyze the output
• The sub-agent will efficiently search and read specific sections using grep and incremental reading

Agent Selection Strategy:

1. If repomix-explorer:explorer agent is available, use it (optimized for repomix output analysis)
2. Otherwise, use the general-purpose agent or another suitable sub-agent
3. The sub-agent should use Grep and Read tools to analyze incrementally

Example:

User: "Pack this codebase and analyze it"

Your workflow:
1. Run: npx repomix@latest
2. Note the output file location (e.g., repomix-output.xml)
3. Launch an appropriate sub-agent with task:
   "Analyze the repomix output file at ./repomix-output.xml.
   Use Grep tool to search for patterns and Read tool to examine specific sections.
   Provide an overview of the codebase structure and main components.
   Do NOT read the entire file at once."

Help and Documentation

If you need more information about available options or encounter any issues:

• Run npx repomix@latest -h or npx repomix@latest --help to see all available options
• Check the official documentation at https://github.com/yamadashy/repomix

Remember: Parse the user's natural language request and translate it into the appropriate repomix command. For analysis tasks, delegate to appropriate sub-agents to avoid consuming excessive context.