Skip to content

Context Tags - A technique for managing context windows in LLM-assisted development. Use tag-based searches to retrieve only the relevant code elements from your project, accurately providing context for LLM-assisted development. Perfect for project enhancements, adding new features, refactors and codebase understanding.

License

Notifications You must be signed in to change notification settings

brandondocusen/ContextTags

Repository files navigation

Context Tags

An efficient technique for managing code context when working with Large Language Models. Context Tags retrieve only the relevant code segments of your code base through tag-based searches, making it easier to provide focused, efficient context during LLM interactions.

Overview

Search for code segments by categorical tags or by granular code element names, retrieve related components, and easily copy the results into LLM chat windows. Context Tags supports multiple approaches to code discovery - from traditional tag matching to vector databases for semantic search and graph databases for relationship exploration.

Features

  • Tag-based code segment retrieval
  • Interactive command line interface
  • Smart tag suggestions
  • Copy-paste friendly output
  • Knowledge graph integration
  • Related code discovery
  1. Search using tags:
  • Enter tags separated by commas (e.g., "String, Logger")
  • Type 'tags' to see all available tags
  • Type 'help' for usage information
  • Type 'quit' or press Ctrl+C to exit

Contributing

Contributions are welcome. Please read CONTRIBUTING.md for guidelines.

Getting Help

  • Review the documentation in the /docs folder
  • Check existing issues or create a new one
  • Read the sample knowledge graph for tag examples

License

MIT License - See LICENSE file for details.


Author: Brandon Docusen
Version: 1.0.0
Status: Beta Release

About

Context Tags - A technique for managing context windows in LLM-assisted development. Use tag-based searches to retrieve only the relevant code elements from your project, accurately providing context for LLM-assisted development. Perfect for project enhancements, adding new features, refactors and codebase understanding.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages