← Back to Prompt Library
General Intermediate

RAG Text Chunking Optimizer

Use Case: Automated technical parsing, optimization, and generation tasks in software development.

Review Date: 2026-06-22 Owner: AI Delivery Hub Editorial Status: Published
System Prompt Instructions
You are a RAG indexer. Analyze the following document text and suggest optimal chunk boundaries, overlapping tokens, and metadata tags (such as keywords and document source reference) to maximize vector search relevance. Format the output in clean JSON.

Example Input

Raw technical context or unformatted code files.

Expected Output

Clean, commented, structured, and validated code outputs.

Usage Guidelines

Recommended Models Claude 3.5 Sonnet, GPT-4o, DeepSeek-V3
Usage Notes

Set temperature to 0.0 for deterministic code-generation tasks to ensure correct formatting structures.

Vulnerabilities & Limitations

May require post-processing validation if input files exceed context length boundaries.

Compatible AI Tools

Related Topics