Intelligent Text Chunking

Documents

intelligent-text-chunking

Intelligent Text Chunking is an Agent Skill for documents published in the majiayu000/claude-skill-registry repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Split large texts into meaningful, AI-optimized chunks while preserving semantic coherence and document structure. Use when processing large documents for AI training, RAG systems, or when you need to break down content while maintaining context and relationships.

View on GitHub ↗

QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

Open your project in Codex.
Copy and paste the prompt below.
Review files and risks before approving any action.

Prompt to paste

Inspect the Agent Skill at https://github.com/majiayu000/claude-skill-registry/blob/HEAD/skills/data/intelligent-text-chunking/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/intelligent-text-chunking, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Intelligent Text Chunking

Overview

A sophisticated text segmentation skill that splits large texts into meaningful, AI-optimized chunks while preserving semantic coherence, document structure, and contextual relationships. Essential for processing large documents for AI training, RAG systems, and memory-constrained applications.

Capabilities

Semantic Awareness: Respects topic boundaries and meaning transitions
Structure Preservation: Maintains document hierarchy and formatting context
Multi-language Support: Handles accented characters and diverse writing systems
Configurable Strategies: Multiple chunking approaches for different use cases
Context Preservation: Intelligent overlap management for context continuity
Quality Optimization: Balances chunk size with content coherence

Chunking Strategies

1. Semantic Chunking

Split text based on meaning and topic boundaries rather than arbitrary size limits.

2. Structural Chunking

Follow document organization (headings, sections, lists) for natural divisions.

3. Fixed-Size Chunking

Create consistent-sized chunks with intelligent boundary selection.

4. Sliding Window Chunking

Create overlapping chunks for enhanced context preservation.

Implementation Examples

Language-Aware Processing

# Multi-language sentence detection
sentence_patterns = {
    'english': ['.', '!', '?', ';'],
    'chuukese': ['.', '!', '?'],
    'general': ['.', '!', '?', ';', '。', '！', '？']
}

def detect_language_patterns(text):
    has_accents = bool(re.search(r'[áéíóúàèìòùāēīōūâêîôû]', text))
    return 'chuukese' if has_accents else 'english'

Basic Usage

from .intelligent_chunker import IntelligentTextChunker, ChunkType

chunker = IntelligentTextChunker(
    max_chunk_size=1024,
    overlap_ratio=0.15,
    preserve_sentences=True
)

chunks = chunker.chunk_document(text, ChunkType.SEMANTIC)

Best Practices

Size Balancing: Balance chunk size with content coherence
Context Preservation: Use appropriate overlap for your use case
Language Awareness: Configure for specific languages when known
Quality Validation: Check chunk quality with sample reviews
Use Case Optimization: Choose strategy based on downstream use

Dependencies

re: Regular expression pattern matching
spacy: Advanced sentence segmentation
nltk: Natural language processing utilities
langdetect: Language identification support