What file formats can the Reference Indexer process?

The skill supports PDF, Word (.docx), Markdown (.md), HTML, and plain text files using libraries like PyPDF2 and BeautifulSoup.

Does this skill support scanned PDF documents?

It primarily handles text-based PDFs. For scanned images, it is recommended to run an OCR tool like ocrmypdf before indexing the file.

Where are the indexed references stored?

Documents and the generated index are stored within the '.agentic_sdlc/references/' directory, organized by category folders.

How do I add a new document to the index?

You can use the '/ref-add {path}' command, which validates the file, extracts text, creates a summary, and updates the RAG corpus.

RAG Reference Indexer

Name: RAG Reference Indexer
Author: arbgjr

byarbgjr

データサイエンスとML

Indexes and manages multi-format reference documents to enhance RAG-based context retrieval during development tasks.

概要

The Reference Indexer skill enables developers to build a robust local knowledge base for Claude by automating the extraction and indexing of external documentation. It supports a variety of formats including PDF, Word, and HTML, converting them into a structured corpus for Retrieval-Augmented Generation (RAG). By categorizing documents into legal, technical, or business folders and generating automatic summaries, it allows Claude to provide more accurate, context-aware answers based on project-specific reference materials and industry standards.

主な機能

Multi-format support for PDF, DOCX, HTML, and Markdown extraction
Integrated search functionality with relevance scoring
Automated document summarization and keyword generation
Automated workflow for chunking and embedding preparation
Categorized indexing for legal, technical, and business documentation
0 GitHub stars

ユースケース

Indexing complex legal frameworks or compliance standards for fintech applications
Managing internal technical specifications and RFCs for large software architectures
Creating a searchable knowledge base of business requirements and user manuals

概要

主な機能

Multi-format support for PDF, DOCX, HTML, and Markdown extraction
Integrated search functionality with relevance scoring
Automated document summarization and keyword generation
Automated workflow for chunking and embedding preparation
Categorized indexing for legal, technical, and business documentation
0 GitHub stars

ユースケース

Indexing complex legal frameworks or compliance standards for fintech applications
Managing internal technical specifications and RFCs for large software architectures
Creating a searchable knowledge base of business requirements and user manuals