skills.rest
sayalinvidia
sayalinvidiaCOMMUNITY
Explore all skills by this author →
GitHub

nemo-retriever

Index multimodal documents and enable cross-document search and extraction.

Updated Apr 16, 2026
Search Tags:#pdf#embeddings#multimodal#retrieval#lancedb#document-processing#vector-search
One-click install
npx skills add https://github.com/sayalinvidia/sayali-skills-test --skill nemo-retriever
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: nemo-retriever
Source: https://github.com/sayalinvidia/sayali-skills-test/tree/main/skills/nemo-retriever
Command: npx skills add https://github.com/sayalinvidia/sayali-skills-test --skill nemo-retriever

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

AI workflows struggle to efficiently search across large, multimodal document collections (PDFs, Word, images, audio, video) and extract structured information; manual grep or single-file reads are brittle and slow.

Core Features & Use Cases

  • Supports indexing multimodal corpora with the NeMo Retriever; enables fast cross-document search, page-level extraction, and extraction of quotes or clauses; works with PDFs, images, Office docs, HTML, TXT, audio, and video.
  • Use cases include legal contracts, research reports, invoices, and audits where you need to locate passages, extract data, or summarize across many documents.
  • Quick-start oriented for developers to bootstrap a retrieval workflow in agent skills.

Quick Start

Ingest your document folder with the default Nemo Retriever ingest flow and run a sample query to retrieve the top hits.

Frequently Asked Questions about nemo-retriever

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

FAQPage Schema
How do I search and extract specific passages across multiple PDF documents?▼

You can index a folder of PDF documents to enable fast cross-document vector search, allowing your agent workflows to locate specific passages and extract page-level information efficiently.

Can I use vector search to locate information in audio and video files?▼

Yes, multimodal vector search supports indexing audio and video files alongside PDFs and images, enabling cross-document retrieval and information extraction across diverse media formats within a large corpus.

What is the best way to build a retrieval-augmented workflow for legal contracts?▼

Building a retrieval-augmented workflow for legal contracts requires indexing the documents with a deterministic ingestion flow to enable fast cross-document search and extract specific quotes, clauses, or data points.

Does this multimodal document retrieval approach work with Office documents and HTML?▼

Yes, multimodal document retrieval supports Office documents, HTML, TXT, PDFs, images, audio, and video, allowing you to index diverse file formats into a unified corpus for cross-document search and extraction.

How do I start indexing a document corpus for fast cross-document search?▼

To start indexing a document corpus for cross-document search, you ingest your target document folder using the default deterministic flow, which processes multimodal files and generates embeddings to power fast agent querying.

Related Skills

Explore More Data & Analytics Skills→
lv7devlv7dev

ai-multimodal

Analyze and generate audio, image, video, and document content with Gemini models.

Community
Advanced
OmidZamaniOmidZamani
120

dspy-rag-pipeline

Build and optimize RAG pipelines with ColBERTv2 in DSPy.

Community
Advanced
zepdbzepdb
43

zeppelin

Index and search text with Zeppelin vector store and BM25 full-text indexing.

Official
Advanced
SpillwaveSolutionsSpillwaveSolutions
14

bm25-search

Search exact terms and phrases in the agent-memory index with BM25 scoring.

Official
Intermediate