chunk

What problem does it solve?

This Skill simplifies the process of breaking down large PDF or Markdown files into smaller, more digestible pieces suitable for Large Language Models (LLMs).

Core Features & Use Cases

• Token-based Chunking: Splits text based on token count for LLM input.
• Page-based Chunking: Preserves document layout by splitting into fixed page groups, ideal for multimodal LLMs.
• PRD Chunking: Hierarchically chunks documents like PRDs, preserving structure.
• Use Case: You have a 50-page PDF report and need to feed its content into an LLM for summarization. This Skill can split it into manageable chunks, either by token count or by groups of pages, ensuring the LLM can process it effectively.

Quick Start

Use the chunk skill to split the attached document 'report.pdf' into token-based chunks.