notoriouslab

notoriouslab / vault-search

Public

Semantic search for Obsidian using local Ollama embeddings

45
8
100% credibility
Found Apr 09, 2026 at 45 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

An Obsidian plugin for local semantic search and note discovery that helps users find related content by meaning and rediscover isolated notes, using a local AI helper.

How It Works

1
🔍 Find a smarter way to search notes

You hear about Vault Search, a simple add-on for your Obsidian notes app that finds notes by what they mean, not just exact words, and helps rediscover forgotten ideas—all privately on your computer.

2
📱 Add it to your notes app

Install the tool right inside Obsidian using the built-in plugin installer or by grabbing a few files from its page.

3
🧠 Connect your local thinking helper

Turn on a free local AI helper on your computer and pick simple thinking models that understand English and Chinese well, so it can grasp the meaning of your notes.

4
📖 Let it scan your notes

Click one button to have it quietly read all your notes once, creating a private map of their ideas that stays on your machine forever.

5
Explore your notes
🔎
Quick search

Type a thought or phrase to instantly see the best matching notes ranked by meaning.

🗺️
Smart discovery

See notes related to what you're reading now, especially forgotten ones without links, to spot your blind spots.

6
✨ Make idea maps

One click turns your search or discovery results into a neat note full of links and previews, helping you organize on your terms.

🎉 Rediscover your knowledge

Your notes come alive with meaningful connections, forgotten gems surface effortlessly, and you feel in control of your growing ideas without anything leaving your computer.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 45 to 45 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is vault-search?

Vault Search brings semantic search to Obsidian vaults using local Ollama embeddings in TypeScript, letting you query notes by meaning rather than keywords. It indexes your markdown files offline, supports synonym expansion, and offers a sidebar panel for search plus a "Discover" tab that highlights isolated "cold" notes related to your active ones. Generate Maps of Content (MOCs) from results or auto-summarize notes with a local LLM for better relevance—all without cloud services or API keys.

Why is it gaining traction?

Unlike cloud-dependent semantic search tools or heavy RAG setups, this runs fully local on 8GB laptops, with bilingual English/Traditional Chinese support via optimized models like qwen3-embedding. Features like hot/cold note tiers surface blind spots in large vaults, chunking for long docs, and one-click MOC export make rediscovery effortless. Incremental indexing and Canvas drag integration hook Obsidian power users tired of manual linking.

Who should use this?

Obsidian users with sprawling vaults full of unlinked research notes or daily journals, especially those handling bilingual content. PKM enthusiasts importing bulk files who want to uncover hidden connections without semantic search Elasticsearch overhead. Developers experimenting with local Ollama for semantic search meaning in personal knowledge bases.

Verdict

Solid early pick for local semantic vault search (45 stars, 1.0% credibility)—docs are thorough, bilingual, MIT-licensed, but low adoption means watch for edge cases. Try via BRAT if your vault needs smarter discovery over keyword hunts.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.