Skip to content

Roadmap

This page describes the near-term direction for Crader.

Current status

Crader uses Tree-sitter for parsing and chunking, enabling true file-incremental indexing. The indexer only processes changed files when re-indexing a repository.

Planned work

  • Extend semantic tagging queries to all supported languages.
  • Preserve the current storage model and retrieval APIs.
  • Add support for additional programming languages.
  • Improve chunking strategies for large files.