skills.rest
ItMeDiaTech
ItMeDiaTechCOMMUNITY
Explore all skills by this author →
GitHub

RAG Retrieval Skill

Semantically query local documents with hybrid retrieval and source citations.

50|10|Updated Oct 30, 2025
Search Tags:#documentation#context#AI assistant#semantic search#RAG#retrieval#local
One-click install
npx skills add https://github.com/ItMeDiaTech/rag-cli --skill rag-retrieval-skill
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: RAG Retrieval Skill
Source: https://github.com/ItMeDiaTech/rag-cli/tree/main/src/plugin/skills/rag-retrieval
Command: npx skills add https://github.com/ItMeDiaTech/rag-cli --skill rag-retrieval-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires sentence-transformers, faiss-cpu, anthropic, langchain, pypdf2, python-docx, and includes scripts (resource) components.

What problem does it solve?

This Skill eliminates the tedious manual work of searching through vast documentation, providing instant, context-aware answers directly from your local knowledge base. It solves the problem of generic AI responses by grounding them in your specific project data, saving you time and reducing cognitive load.

Core Features & Use Cases

  • Context-Aware AI Answers: Get precise responses from Claude Haiku, directly informed by your internal documentation, code, or research materials.
  • Hybrid Search Accuracy: Combines semantic understanding with keyword matching to ensure the most relevant documents are retrieved, even for complex queries.
  • Local-First Privacy: All document processing and retrieval occurs on your machine, guaranteeing your sensitive data remains private and secure.
  • Use Case: Ask "How do I configure authentication for the new service?" and receive an answer tailored to your project's specific setup, complete with citations from your internal API documentation, all within seconds.

Quick Start

Ask the RAG Retrieval Skill: "How to configure the API?" To retrieve more context, ask: "How to handle errors?" --top-k 10

Frequently Asked Questions about RAG Retrieval Skill

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

FAQPage Schema
How do I query local documents with semantic search instead of keyword matching?▼

Semantic search retrieves documents by meaning rather than exact keywords. This Skill uses sentence-transformers embeddings and faiss vector indexing to find contextually relevant documents, then sends them to Claude for context-aware answers. Hybrid search combines both methods for higher accuracy.

Can I use RAG to answer questions from my internal documentation without uploading to the cloud?▼

Yes. All document processing, embedding, and retrieval happens locally on your machine using faiss and sentence-transformers. Your sensitive data never leaves your system, ensuring privacy while enabling instant answers grounded in your specific knowledge base.

How do I configure retrieval accuracy for my documentation queries?▼

Configure top-k (number of documents retrieved), similarity threshold (minimum relevance score), and retrieval mode (hybrid, vector, or keyword-only). This Skill handles PDF, DOCX, and other document formats, letting you tune precision and recall for your use case.

What file formats does this Skill support for indexing?▼

This Skill indexes PDF and DOCX documents using pypdf2 and python-docx. It extracts text, converts it to embeddings, and stores vectors in a faiss index for fast semantic retrieval alongside keyword search.

Can I see which documents the AI used to answer my question?▼

Yes. This Skill returns source citations showing which indexed documents informed each answer, plus retrieval latency metrics. This transparency helps verify accuracy and trace information back to your original documentation.

Do I need coding experience to set up local document indexing?▼

This Skill provides scripts for configuration, indexing, and querying. Basic familiarity with Python, file paths, and vector store concepts helps, but the setup automates embedding generation, faiss index creation, and retrieval pipeline wiring.

Related Skills

Explore More Software Engineering Skills→
adeboldadebold

AgentDB Vector Search

Perform vector-based semantic search against an AgentDB index.

Community
Advanced
huangrt01huangrt01
4.0k

search-web

Perform web searches and return answers with cited sources.

Community
Intermediate
lacetolaceto
1

langchain-rag

Build end-to-end RAG pipelines with LangChain document loaders, embeddings, and vector stores.

Community
Advanced
icyzhicyzh
1

qmd

Index and search local markdown notes using BM25 and semantic embeddings.

Community
Advanced