skills.rest
Razaib-khan
Razaib-khanCOMMUNITY
Explore all skills by this author →
GitHub

building-rag-systems

Build Qdrant-backed RAG pipelines with semantic chunking and incremental indexing.

Updated Jan 15, 2026
Search Tags:#rag#qdrant#semantic chunking#incremental indexing#context expansion#filtered retrieval
One-click install
npx skills add https://github.com/Razaib-khan/Hackathon-2-phases-2-3-4-5- --skill building-rag-systems-razaib-khan
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: building-rag-systems
Source: https://github.com/Razaib-khan/Hackathon-2-phases-2-3-4-5-/tree/main/.claude/skills/building-rag-systems
Command: npx skills add https://github.com/Razaib-khan/Hackathon-2-phases-2-3-4-5- --skill building-rag-systems-razaib-khan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires qdrant-client, openai, pydantic, python-frontmatter, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps you build production-ready retrieval-augmented generation systems that can index documents incrementally, search with precise filters, and return surrounding context reliably.

Core Features & Use Cases

  • Semantic Chunking: Split documents on meaningful section boundaries instead of fixed-size text windows.
  • Incremental Indexing: Detect new, modified, and deleted files by hashing content so only changed documents are reprocessed.
  • Filtered Retrieval: Search with tenant isolation, chapter or lesson ranges, proficiency levels, and other indexed metadata.
  • Context Expansion: Walk previous and next chunk links to reconstruct surrounding passages for better answers.
  • Use Case: Build a knowledge base that updates automatically when source docs change and only surfaces results relevant to the current user, module, or lesson.

Quick Start

Use the building-rag-systems skill to design a Qdrant-backed ingestion and retrieval pipeline with semantic chunking, payload indexes, and context-aware search.

Frequently Asked Questions about building-rag-systems

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

FAQPage Schema
How do I build a RAG system with semantic chunking and Qdrant?▼

Build a production RAG pipeline by applying semantic chunking to split documents on section boundaries, then indexing them in Qdrant with payload indexes for filtered vector retrieval.

How does incremental indexing work for RAG document ingestion?▼

Incremental indexing detects new, modified, and deleted files by hashing content. This ensures only changed documents are reprocessed, keeping your knowledge base updated efficiently without full reindexing.

Can I implement tenant-isolated search in a RAG pipeline using Qdrant?▼

Yes, filtered retrieval in Qdrant supports tenant isolation. You achieve this by applying payload indexes to metadata fields, enabling searches restricted to specific users, modules, or lessons.

What is context expansion in retrieval-augmented generation?▼

Context expansion in RAG reconstructs surrounding passages by walking previous and next chunk links. This chain-based neighbor retrieval provides better surrounding context for generating accurate answers.

Do I need OpenAI and Pydantic to set up filtered vector retrieval?▼

Yes, the pipeline uses OpenAI for embeddings and Pydantic for data validation. These dependencies work with the Qdrant client to manage semantic chunking and payload-based filtered retrieval.

What's the best way to update a knowledge base when source documents change?▼

The best way is using a pipeline with file-hash change detection. This incremental indexing approach automatically identifies modified documents and only reprocesses the changed files in your Qdrant collection.

Related Skills

Explore More Software Engineering Skills→
JoranHonigJoranHonig
63

librarian-index

Builds a deterministic semantic vector index for registered library files using FastEmbed and Qdrant.

Community
Intermediate
floflo777floflo777
33

rag-eval

Evaluate RAG systems using retrieval, generation, and latency metrics.

Community
Intermediate
mikkelkrogsholmmikkelkrogsholm

bookstrap-ingest

Ingest PDFs, Markdown, HTML, and plain text into Bookstrap knowledge bases with semantic chunking and embeddings.

Community
Advanced
AbdullahMalik17AbdullahMalik17
9

building-rag-systems

Build production RAG systems with semantic chunking and Qdrant vector search.

Community
Advanced