AST-chunked languages
These languages use tree-sitter for semantic chunking — functions, classes, and methods are extracted as individual chunks:- Python
- JavaScript / TypeScript
- Rust
- Go
- Java
- C / C++
- Ruby
- Swift
- Kotlin
- C#
Fallback: sliding window
All other text files (.txt, .md, config files, etc.) fall back to sliding-window line chunks, ensuring no file is silently skipped.