How does it handle scanned PDF documents?

While it defaults to standard text extraction via pdftotext or PyPDF2, it can be integrated with OCR tools like ocrmypdf to process scanned image-based documents.

How does it integrate with the RAG corpus?

Indexed documents are processed into chunks, enriched with metadata and summaries, and then converted into embeddings within the .agentic_sdlc/corpus for AI retrieval.

What file formats does the Reference Indexer support?

The skill supports a wide range of formats including PDF, Word (.docx), Markdown (.md), Plain Text (.txt), and HTML.

Can I organize documents into different categories?

Yes, the skill uses a structured directory system to categorize documents into legal, technical, business, and internal folders for better organization.

Reference Indexer & RAG Manager

Name: Reference Indexer & RAG Manager
Author: arbgjr

byarbgjr

Data Science & ML

Indexes and manages local documents like PDFs and Word files to enhance Claude's knowledge through Retrieval-Augmented Generation (RAG).

About

The Reference Indexer skill transforms Claude into a domain-specific expert by systematically indexing external documentation, including legal statutes, technical specifications, and business manuals. It automates text extraction from various formats such as PDF, DOCX, HTML, and Markdown, generating summaries and keywords while organizing files into a structured, searchable corpus. This allows Claude to provide highly accurate, context-aware answers based on your private or specialized reference library, ensuring that project-specific information is always accessible during the development process.

Key Features

Full-text and semantic search capabilities within the indexed corpus
Automatic document summarization and keyword extraction
Integrated RAG workflow with customizable chunking and metadata management
Structured organization for legal, technical, and business references
Multi-format text extraction for PDF, DOCX, MD, and HTML files
0 GitHub stars

Use Cases

Managing business manuals and internal documentation for streamlined team onboarding
Building a searchable legal or regulatory compliance knowledge base for software projects
Indexing technical RFCs and internal specifications for context-aware code generation

About

Key Features

Full-text and semantic search capabilities within the indexed corpus
Automatic document summarization and keyword extraction
Integrated RAG workflow with customizable chunking and metadata management
Structured organization for legal, technical, and business references
Multi-format text extraction for PDF, DOCX, MD, and HTML files
0 GitHub stars

Use Cases

Managing business manuals and internal documentation for streamlined team onboarding
Building a searchable legal or regulatory compliance knowledge base for software projects
Indexing technical RFCs and internal specifications for context-aware code generation