skills.rest
Gherkin
GherkinCOMMUNITY
Explore all skills by this author →
GitHub

pdf-rag-knowledge

Search a local PDF knowledge base to answer technical documentation questions with citations.

1|Updated Dec 25, 2025
Search Tags:#ollama#semantic-search#retrieval-augmented-generation#vector-store#datasheets#pdf-rag#hardware-docs
One-click install
npx skills add https://github.com/Gherkin/vscode-pdf-rag-skill --skill pdf-rag-knowledge
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pdf-rag-knowledge
Source: https://github.com/Gherkin/vscode-pdf-rag-skill/tree/main
Command: npx skills add https://github.com/Gherkin/vscode-pdf-rag-skill --skill pdf-rag-knowledge

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, PyPDF2, pycryptodome, and includes scripts (resource) components.

What problem does it solve?

It solves the problem of getting accurate answers from hardware manuals and datasheets without manually reading and searching through PDFs.

Core Features & Use Cases

  • Semantic search over indexed PDFs: Retrieves the most relevant chunks using embeddings and cosine similarity.
  • Local, repo-scoped knowledge base: Stores embeddings in a local vector_store.json for portability and privacy.
  • Ollama-based embedding generation: Uses a locally running Ollama model to embed text chunks, then supports quick question answering with citations (source and page).

Quick Start

Index your PDFs by running python3 rag_search.py --index /path/to/your/datasheets/*.pdf.

Frequently Asked Questions about pdf-rag-knowledge

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

FAQPage Schema
How do I search PDF datasheets for technical specs without reading the whole document?▼

You can search PDF datasheets using semantic search to retrieve relevant text chunks. A local RAG system indexes documents by generating embeddings, returning matching sections with source citations instead of requiring manual reading.

Can I use Ollama to build a local retrieval augmented generation pipeline for hardware manuals?▼

Yes, you can build a local retrieval augmented generation pipeline for hardware manuals using Ollama. It serves as the embedding model endpoint to process text chunks into vectors stored in a local JSON file.

How do I index IC datasheets and FPGA manuals for question answering with source citations?▼

To index IC datasheets and FPGA manuals, run a script that parses PDFs into chunks, generates embeddings via Ollama, and stores them in a local vector store. Queries then return matching chunks with source filename and page citations.

Does semantic search over technical PDFs work offline without sending data to external APIs?▼

Semantic search over technical PDFs works offline by using a local Ollama endpoint for embeddings and storing vectors in a repo-local JSON file. This ensures data privacy without relying on external APIs.

What are the limitations of using a local vector store JSON for retrieval augmented generation?▼

Limitations of using a local vector store JSON include potential performance bottlenecks with large-scale PDF libraries. It is designed for repo-scoped portability and privacy rather than distributed enterprise-level concurrent queries.

Related Skills

Explore More Software Engineering Skills→
ScavieFaeScavieFae
9

setup

Detect the host platform, install dependencies, and configure Dolphin, Slippi, and Phillip.

Community
Advanced
zeenie-aizeenie-ai
356

perplexity-search-skill

Answer questions with synthesized web research and verifiable citations via Perplexity Sonar.

Community
Intermediate
ericwang915ericwang915
38

pdf_split

Extract specified pages from PDFs into new files using PyPDF2.

Community
Intermediate
davpatel605-beepdavpatel605-beep

chroma

Store and retrieve vector embeddings for semantic search and document retrieval.

Community
Intermediate