Compress structured data into TOON format to reduce LLM token usage.
npx skills add https://github.com/modu-ai/moai-adk --skill moai-formats-data
Please help me install this Agent Skill. Skill: moai-formats-data Source: https://github.com/modu-ai/moai-adk/tree/main/.claude/skills/moai-formats-data Command: npx skills add https://github.com/modu-ai/moai-adk --skill moai-formats-data
Data format management for modern apps, covering TOON encoding, JSON/YAML optimization, serialization patterns, and data validation to reduce payload size and improve performance.
Run a quick TOON encode/decode: encoder = TOONEncoder(); toon = encoder.encode({"a":1}); original = encoder.decode(toon)
High-intent search queries and answers about installing and using this skill.
TOON encoding achieves 40-60% token reduction on typical structured data compared to JSON, making it ideal for LLM communication where payload size directly impacts cost and latency.
Use the TOONEncoder to convert between formats: instantiate the encoder, call encode() to convert JSON/YAML to TOON, and decode() to restore the original data losslessly.
Yes, the Skill includes streaming and batch validation with schema management, enabling type checks and format validation during serialization for performance-critical workflows.
The Skill supports streaming large datasets and dynamic schema evolution, allowing you to encode and validate data in batches without loading entire payloads into memory.
TOON encoding and validation run on Python 3.8 and later, with optional orjson integration for ultra-fast JSON handling on supported platforms.
walter-laExtract structured data from unstructured text using a shared JSON schema.
hnabyz-bot...
ConsiliencyEncode JSON structures into compact TOON format for LLM prompts.
AGIBuildGuide .NET serialization format selection and implementation for AOT compatibility.
