How do I start using the RAG system?

First, run /memvid-rag:create to initialize your index, then use /memvid-rag:index to add documents, and finally /memvid-rag:search to query them.

What file formats does Memvid Local RAG support?

It supports PDF by default, and can also index DOCX and XLSX files if the python-docx and openpyxl libraries are installed via your Python environment.

Where are my indexed documents stored?

All indexed data and embeddings are stored locally in a file named knowledge.mv2 within your current working directory.

Do I need an internet connection for this skill?

While your data is stored locally, an internet connection and an OpenAI API key are required to generate the vector embeddings used for semantic search.

Can I filter search results by relevancy?

Yes, the search command supports a --min_relevancy parameter to ensure only the most contextually appropriate snippets are returned to Claude.

Memvid Local RAG

Name: Memvid Local RAG
Author: fa-ina-tic

byfa-ina-tic

•

Data Science & ML

Enables local semantic search and document indexing for PDF, DOCX, and XLSX files directly within Claude Code.

Memvid Local RAG empowers Claude Code with a local Retrieval-Augmented Generation (RAG) system, allowing you to build and query a private knowledge base from your own documents. By indexing files like PDFs and spreadsheets into a local vector database, this skill provides semantic search capabilities that ground Claude's responses in your specific technical documentation, research papers, or project data without needing external cloud hosting for your knowledge base. It is an essential tool for developers who need to reference large volumes of local documentation while maintaining data privacy.

Key Features

01Adjustable relevancy thresholds and snippet lengths for precise retrieval

02Customizable search modes including lexical, semantic, and auto

03Local semantic search across PDF, DOCX, and XLSX formats

042 GitHub stars

05Privacy-focused local storage in a knowledge.mv2 file

06Vector-based document indexing using OpenAI embeddings

Use Cases

01Grounding coding assistance in proprietary technical documentation and API specs

02Searching across large research papers to extract specific architectural patterns

03Indexing project requirements and spreadsheets for context-aware code generation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add fa-ina-tic/memvid-rag skills

For use in Claude.ai and ChatGPT

Download Skill