bmad-distillator

Yes, you can compress large text documents into dense distillates without losing factual information. The Skill performs lossless compression by removing unnecessary text and redundancy while retaining all decisions, constraints, and relationships.

Provide your source file to the Skill, which analyzes the content and performs semantic splitting based on a token budget. It then generates a single, dense distillate document optimized for LLM consumption.

Yes, the compression process evaluates the document's content and performs semantic splitting based on a defined token budget. This ensures the final dense distillate remains optimized for efficient LLM consumption.

Unlike standard summarization, lossless document compression preserves all factual information, decisions, and relationships while removing only unnecessary text. This ensures the dense output remains fully reconstructable for LLM information extraction.

Use document compression when you have large files containing detailed information that an LLM needs to process quickly. It prevents information overload and context window limits by creating a dense, redundancy-free distillate for decision-making.