Our review
Reads documents and splits them into manageable chunks for processing, analysis, or transfer.
Strengths
- Multiple splitting strategies (lines, size, sections, paragraphs)
- Supports overlap to maintain context between chunks
- Clear output with headers and summaries
- Simple usage with intuitive options
Limitations
- Only works on readable text files
- Size-based splitting is approximate and breaks at line boundaries
- Relies on line break or header structure
When you need to break a large document into smaller pieces for sequential processing or analysis.
For binary files or when byte-precise splitting is required.
Security analysis
SafeThe skill only reads files and uses 'wc -l' for counting lines, with no network access, file modification, or command execution beyond the allowed tools. It presents no risk of data exfiltration, unauthorized access, or destructive actions.
No concerns found
Examples
/split README.md/split document.md --sections/split document.txt --lines 100 --overlap 10name: split description: Read documents and split them into chunks allowed-tools:
- Read
- Bash(wc *)
Split Skill
Read documents and split them into manageable chunks for processing, analysis, or transfer. Like a master breaking boards, this skill shatters overwhelming information into fragments that can be easily mastered.
Capabilities
- Split by Lines: Split files into chunks of a precise number of lines.
- Split by Size: Split files into chunks of an approximate character count.
- Split by Sections: Split at structural boundaries like markdown headers.
- Split by Paragraphs: Split at blank line delimiters.
- Support Overlap: Maintain context between chunks with an overlapping buffer.
Usage Examples
Split a file into standard chunks:
/split README.md
Split by precise line count:
/split document.txt --lines 100
Split by markdown sections:
/split document.md --sections
Split by paragraphs:
/split article.txt --paragraphs
Split with context overlap:
/split document.txt --lines 100 --overlap 10
Options
| Option | Description | Default |
| -------------- | ------------------------------------ | ------- |
| --lines N | Split every N lines | 100 |
| --size N | Split at approximately N characters | - |
| --sections | Split at markdown headers (# ## ###) | - |
| --paragraphs | Split at blank lines | - |
| --overlap N | Include N lines of context | 0 |
Output Format
Present each chunk with a header showing its number and line range:
══════════════════════════════════════════════════════
CHUNK 1 of 5 (lines 1-100)
══════════════════════════════════════════════════════
[chunk content here]
══════════════════════════════════════════════════════
CHUNK 2 of 5 (lines 101-200)
══════════════════════════════════════════════════════
[chunk content here]
Splitting Strategies
By Lines (default)
Split at fixed line intervals. Best for uniform content like logs or data files.
By Size
Split at approximate character counts, breaking at line boundaries. Best when targeting specific output sizes.
By Sections
Split at markdown headers (lines starting with #). Best for documentation. Each section becomes its own chunk.
By Paragraphs
Split at blank lines. Best for prose where paragraphs are the logical units.
Workflow
- Read the file using the Read tool.
- Get file stats using
wc -lto show total lines. - Apply splitting strategy based on options.
- Display chunks with headers and summaries.
- Show summary with total chunks and average chunk size.
Prompt Engineering
Data & AI
Prompt engineering best practices and templates to maximize AI outputs.
Data Visualization
Data & AI
Generates data visualizations and charts tailored to your data.
RAG Architecture Setup
Data & AI
Setup guide for RAG (Retrieval-Augmented Generation) architectures.