toon-formatter

TOON v2.0 compresses structured data by converting arrays, tables, and API responses into a compact encoded format. It detects suitable datasets (5+ items, 60%+ field uniformity) and applies tabular, inline, or expanded array encoding with configurable delimiters to minimize token count while preserving data integrity.

TOON v2.0 handles arrays, tables, logs, events, transactions, analytics records, API responses, and database results. It automatically detects structured data with uniform fields and supports three array types (tabular, inline, expanded) and three delimiters (comma, tab, pipe) to fit various output shapes.

Yes. TOON v2.0 includes key folding to flatten nested objects, maximizing token savings by collapsing hierarchical data into a single level before encoding. This is especially useful for complex API responses and database results with nested fields.

Use TOON v2.0 when handling arrays with at least 5 items and consistent field structure (60%+ uniformity). It's most effective in agent workflows, tool calls, RAG systems, and batch processing where token efficiency directly reduces latency and cost without sacrificing data accuracy.

TOON v2.0 offers three array encoding types optimized for different data shapes: tabular for dense columnar data, inline for compact single-line output, and expanded for hierarchical or sparse records. Choose based on your data structure and delimiter preference (comma, tab, or pipe).