Use-case shortlist
Find skills for document ingestion, retrieval, embeddings, source-grounded answers, and agent workflows that need reliable private knowledge.
Decision prompt
I need my agent to build a RAG workflow over documents, retrieve reliable context, and answer with grounded sources.
Recommended shortlist
This repository showcases various advanced techniques for Retrieval-Augmented Generation (RAG) systems. Each technique has a detailed notebook tutorial.
$ npx skills add NirDiamant/RAG_Techniques💡 All-in-one AI framework for semantic search, LLM orchestration and language model workflows
$ npx skills add neuml/txtaiMilvus is a high-performance, cloud-native vector database built for scalable vector ANN search
$ npx skills add milvus-io/milvus📑 PageIndex: Document Index for Vectorless, Reasoning-based RAG
$ npx skills add VectifyAI/PageIndexHow to use this guide
Pick a small but representative set of documents before scaling ingestion.
Ask known-answer questions and inspect whether the right source material appears.
Require the agent to show the evidence behind each answer before shipping.
Evaluation notes
The point of a RAG skill is not only retrieval. It should help an agent ingest clean material, retrieve relevant context, and keep answers grounded in sources.
Many failures happen before retrieval. Web scraping, PDF parsing, OCR, and document cleanup skills often matter as much as the RAG layer.
FAQ
If the source material lives on the web, a scraping skill can be the upstream ingestion layer. For private files, document processing may be more important.
Start with a skill that can ingest and retrieve from your actual source format, then add companions for parsing, crawling, or evaluation.
More candidates
Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/
LLPhant - A comprehensive PHP Generative AI Framework using OpenAI GPT 4. Inspired by Langchain
[MLsys2026]: RAG on Everything with LEANN. Enjoy 97% storage savings while running a fast, accurate, and 100% private RAG application on your personal device.
A multi-modal vector database that supports upserts and vector queries using unified SQL (MySQL-Compatible) on structured and unstructured data, while meeting the requirements of high concurrency and ultra-low latency.
Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.
Distributed vector search for AI-native applications
A simple, fast and versatile Datalog database
Langchain-Chatchat（原Langchain-ChatGLM）基于 Langchain 与 ChatGLM, Qwen 与 Llama 等语言模型的 RAG 与 Agent 应用 | Langchain-Chatchat (formerly langchain-ChatGLM), local knowledge based LLM (like ChatGLM, Qwen and Llama) RAG and Agent app with langchain
Next guides
Use-case shortlist
Compare skills for crawling sites, extracting structured data, converting pages to markdown, and feeding reliable web context into agent workflows.
Platform shortlist
A focused guide for builders using Codex-style coding agents: repository inspection, issue triage, implementation planning, testing, and browser verification skills.
Installation guide
A step-by-step workflow for adding install-ready skills to a Codex-style agent setup while keeping validation, permissions, and rollback in mind.