Agent skill · data analytics · aiskillstore
building-rag-systems
Build production RAG systems with semantic chunking, incremental indexing, and filtered retrieval. Use when implementing document ingestion pipelines, vector search with Qdrant, or context-aware retrieval. Covers chunking strategies, change detection, payload indexing, and context expansion. NOT when doing simple similarity search without production requirements.
Why this skill is useful
Adds executable scripts and detailed architecture for building production-grade RAG systems with specific chunking and indexing strategies.
What it needs
Requires openai, pydantic, python-frontmatter, qdrant-client installed locally. About 4k tokens when loaded. Last updated 2026-08-07. 405 stars on the source repository.
What this skill does
Building RAG Systems Production-grade RAG with semantic chunking, incremental updates, and filtered retrieval. Quick Start --- Ingestion Pipeline Architecture Semantic Chunking (NOT Fixed-Size) Change Detection (Incremental Updates) Batched Embeddings Qdrant Collection with Payload Indexes --- Retrieval Patterns Comprehensive Filter Builder Context Expansion (Walk Chunk Chain) Full Document Retrieval --- Payload Schema --- Anti-Patterns Don't Do Instead ------- ------------ Fixed character chunking Semantic boundaries (## headers) Position-based chunk IDs Content hash for stable IDs No overlap between chunks 10-20% overlap for continuity Full re-index on every change Incremental with file hash detection Missing payload indexes Index every field you filter by Synchronous embedding Batch with background jobs External state database Qdrant-native state tracking --- Verification Run: python scripts/verify.py Related Skills scaffolding-fastapi-dapr - API patterns for search endpoints streaming-llm-responses - Streaming RAG responses References references/ingestion-patterns.md - Full ingestion pipeline references/retrieval-patterns.md - Filter strategies, context expansion
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills aiskillstore/building-rag-systems