skills.rest
Postalocity
PostalocityOFFICIAL
Explore all skills by this author →
GitHub

rag-engineer

Design and optimize RAG pipelines with embeddings, chunking, and vector stores.

1|Updated Mar 5, 2026
Search Tags:#llm#rag#chunking#embedding#vector-database#retrieval-augmented#hybrid-search
One-click install
npx skills add https://github.com/Postalocity/template-microsite --skill rag-engineer-postalocity
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rag-engineer
Source: https://github.com/Postalocity/template-microsite/tree/main/.opencode/integrations/rag-engineer
Command: npx skills add https://github.com/Postalocity/template-microsite --skill rag-engineer-postalocity

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Retrieval-Augmented Generation (RAG) systems often suffer from hallucinations and context drift when relying on large text corpora. This Skill provides a blueprint to design, optimize, and operate robust retrieval pipelines that maintain accuracy and relevance across documents and knowledge bases.

Core Features & Use Cases

  • Vector embeddings and similarity search
  • Document chunking and preprocessing
  • Retrieval pipeline design
  • Semantic search implementation
  • Context window optimization
  • Hybrid search (keyword + semantic)
  • Use Case: enterprise knowledge bases, customer support agents, research assistants

Quick Start

Design and implement a retrieval-augmented generation pipeline for a given corpus.

Frequently Asked Questions about rag-engineer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a RAG pipeline to reduce LLM hallucinations on enterprise data?▼

Build a retrieval-augmented generation pipeline using document chunking, vector embeddings, and similarity search to ground LLM responses. This improves factual accuracy by retrieving relevant context from your knowledge base before generation.

What is the best way to chunk documents for a retrieval-augmented generation system?▼

Document chunking for RAG involves splitting text at semantic boundaries to preserve context. Optimizing chunk boundaries ensures the vector database stores coherent information, leading to more precise semantic search results.

Does hybrid search improve retrieval accuracy over semantic search alone?▼

Hybrid search combines keyword matching with semantic similarity to improve retrieval accuracy. Integrating both methods in your RAG pipeline ensures precise retrieval of specific terms while maintaining contextual relevance.

How do I manage context windows when retrieving from a vector database?▼

Context window management in RAG limits the number of retrieved chunks passed to the LLM. Optimizing this prevents token overflow and ensures the generation model focuses on the most relevant retrieved context.

Can I use this approach for customer support agents and research assistants?▼

Yes, retrieval-augmented generation pipelines suit enterprise knowledge bases, customer support agents, and research assistants. The system applies semantic search and context grounding to deliver accurate responses across these applications.

Why does my RAG system suffer from context drift on large text corpora?▼

RAG systems experience context drift when retrieval fails to find relevant information in large corpora. Applying robust error handling, precise chunking strategies, and hybrid search integration maintains accuracy and relevance.

Related Skills

Explore More Software Engineering Skills→
jolequesjoleques

documentador-rag

Convert unstructured inputs into RAG-ready Markdown documents with YAML metadata.

Community
Intermediate
MantisWareMantisWare

eval-rag

Evaluate RAG pipelines with retrieval and generation metrics and bottleneck analysis.

Community
Advanced
melikhanmutlumelikhanmutlu

rag-implementation

Build retrieval-augmented generation pipelines with vector databases and semantic search.

Community
Advanced
KarlinskySKarlinskyS

unsloth

Fine-tune large language models with Unsloth using LoRA/QLoRA adapters.

Community
Advanced