bmad-distillator

Compress source documents into lossless, token-efficient distillates for LLM workflows.

1|Updated Dec 22, 2022
One-click install
npx skills add https://github.com/Rinzler78/osmosis-launcher --skill bmad-distillator-rinzler78
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-distillator
Source: https://github.com/Rinzler78/osmosis-launcher/tree/main/.agents/skills/bmad-distillator
Command: npx skills add https://github.com/Rinzler78/osmosis-launcher --skill bmad-distillator-rinzler78

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill turns one or more source documents into a dense, token-efficient distillate that preserves facts, decisions, constraints, relationships, and named entities while stripping filler and repetition. It is designed for workflows that need the original information in a much smaller context window without losing critical details.

Core Features & Use Cases

  • Lossless document compression: Extracts and deduplicates discrete facts, requirements, decisions, risks, and open questions from source files.
  • Semantic grouping and splitting: Organizes content into coherent themes and creates split distillates when the material is too large for a single file.
  • Validation workflow: Supports completeness checks and optional round-trip reconstruction testing to detect gaps.
  • Use cases: Distilling product briefs, architecture notes, research docs, or multi-file document sets for downstream LLM analysis, PRD creation, or architecture review.

Quick Start

Ask the skill to distill the provided source documents into a single lossless markdown distillate optimized for downstream LLM use.

Frequently Asked Questions about bmad-distillator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I compress multiple documents into a lossless distillate for LLM context?

Document distillation compresses source files into a token-efficient distillate that preserves facts, decisions, constraints, and named entities. It extracts and deduplicates discrete information while stripping filler and repetition for downstream LLM workflows.

What is the best way to extract knowledge from large documents when context is too large for one file?

Semantic splitting organizes content into coherent themes and creates split distillates when material exceeds a single file's context window. This ensures thematic consolidation preserves critical details without overwhelming the LLM context limit.

Can I distill architecture notes and research docs for downstream PRD creation?

Yes, distilling product briefs, architecture notes, and research docs produces a dense markdown distillate optimized for downstream analysis. It preserves requirements, risks, and open questions for PRD creation or architecture review.

Does document distillation preserve named entities and dependencies from the source files?

Document distillation explicitly preserves facts, decisions, constraints, dependencies, risks, and named entities while removing filler. It maintains these critical elements during semantic grouping to ensure lossless compression.

What are the limitations of semantic splitting for multi-file document analysis?

Semantic splitting requires coherent thematic grouping to function effectively, meaning highly fragmented or unstructured source documents may yield less optimal distillates. Completeness checks are recommended to validate the split distillates.