agentsclimarketplace
Sign in

Chunking

Skill Goodnight77/rag-skills/skills/chunking

Route RAG chunking decisions across semantic, hierarchical, sliding-window, contextual-header, and framework-selection strategies.From its SKILL.md

Install
npx -y skills add Goodnight77/rag-skills --skill chunking

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 7 stars7 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.5 KB, as published. Nobody here has run it

Chunking

Overview

Use this parent skill when the main RAG problem is how to split source material into retrievable units. Route to the child skill that best matches the document shape, retrieval failure, and production constraints.

Problem Statement

Poor chunking causes retrieval misses, fragmented answers, duplicated context, and weak citations. RAG systems need chunking strategies that preserve meaning while staying efficient for indexing and retrieval.

Key Concepts

  • Semantic boundaries: Split where ideas naturally change.
  • Hierarchy: Preserve document, section, and subsection relationships.
  • Overlap: Add controlled neighboring context across chunk boundaries.
  • Context headers: Attach inherited titles and metadata to chunks.

Implementation Guide

Step 1: Identify the Corpus Shape

Use semantic chunking for prose, hierarchical chunking for structured documents, and code-specific guidance for APIs or source repositories.

Step 2: Match the Retrieval Failure

Use sliding windows when answers cross boundaries, contextual headers when chunks lack source context, and semantic chunking when fixed windows split ideas.

Step 3: Validate Chunk Quality

Measure retrieval recall, answer groundedness, duplicate context rate, and citation usefulness before changing embeddings or vector databases.

When to Use This Skill

  • Choosing a chunking strategy for a new RAG pipeline
  • Debugging poor recall caused by fragmented source content
  • Improving citations and source traceability

When NOT to Use This Skill

  • The bottleneck is ranking after retrieval
  • The issue is vector database operations or deployment
  • Source documents are already chunked and validated

Related Skills

Metrics & Success Criteria

  • Higher recall on chunk-boundary questions
  • Lower duplicate or irrelevant context in prompts
  • Better source citations and more coherent retrieved passages

Keep looking

Skills are one crate of 627,801. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.