skills.rest
arunoda
arunodaCOMMUNITY
Explore all skills by this author →
Website𝕏@arunodaGitHub

RAG Fact Search

Search extracted facts locally using Ollama embeddings and sqlite-vec.

21|4|Updated Feb 20, 2026
Search Tags:#semantic search#rag#knowledge base#ollama#sqlite-vec#fact retrieval
One-click install
npx skills add https://github.com/arunoda/OkBrain --skill rag-fact-search
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: RAG Fact Search
Source: https://github.com/arunoda/OkBrain/tree/main/.agent/skills/rag_fact_search
Command: npx skills add https://github.com/arunoda/OkBrain --skill rag-fact-search

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill enables efficient retrieval of specific, extracted facts from your personal knowledge base using semantic search, ensuring AI responses are grounded in your data.

Core Features & Use Cases

  • Local Semantic Search: Utilizes Ollama embeddings and sqlite-vec for fast, private vector search over extracted facts.
  • Fact Retrieval: Finds relevant facts based on the semantic meaning of your query, not just keywords.
  • Use Case: When asking a question, the AI can automatically find and use previously extracted facts about a specific topic (e.g., "What were the key takeaways from my meeting about Project X?") to provide a more informed and personalized answer.

Quick Start

Ask the AI a question about a topic you've previously discussed to see relevant facts automatically included in the response.

Frequently Asked Questions about RAG Fact Search

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

FAQPage Schema
How do I perform local semantic search over extracted facts?▼

You can perform local semantic search over extracted facts by using Ollama to generate query embeddings and sqlite-vec to handle vector storage and retrieval. This allows the AI to find relevant facts based on semantic meaning rather than exact keywords.

Do I need Ollama and sqlite-vec to retrieve facts for AI context injection?▼

Yes, retrieving facts for AI context injection with this Skill requires Ollama to compute the embeddings from your queries and sqlite-vec to store and search the vector data locally.

What is the best way to find relevant facts from my personal knowledge base?▼

The best way to find relevant facts is using local semantic search, which matches the meaning of your query against extracted facts stored in a vector database. This ensures your AI responses are grounded in your specific data.

Can I use sqlite-vec for private vector search over my own data?▼

Yes, you can use sqlite-vec for private vector search over your own data. It stores and queries vector embeddings locally, ensuring your personal knowledge base and fact retrieval process remain completely private.

How does semantic fact retrieval work without relying on keyword matching?▼

Semantic fact retrieval works by converting your query into embeddings using Ollama and comparing them against stored vectors in sqlite-vec. This process matches the underlying meaning of your question to relevant facts, rather than relying on exact keyword matching.

Related Skills

Explore More Software Engineering Skills→
dandakadandaka
111

traul

Index Slack, Telegram, Discord, Linear, Gmail, and WhatsApp messages locally for unified search.

Community
Advanced
thecloudtipsthecloudtips
18

Answer Synthesis

Synthesize sourced answers with inline citations and confidence ratings.

Community
Advanced
anderskevanderskev
20

sqlite-vec

Store and query vector embeddings in SQLite using vec0 virtual tables.

Community
Intermediate
KappTech88KappTech88

chroma

Store and retrieve vector embeddings with metadata for semantic search.

Community
Advanced