scambier/obsidian-omnisearch
A search engine that "just works" for Obsidian. Supports OCR and PDF indexing.
What it solves
Omnisearch addresses the difficulty of quickly locating specific information across a large Obsidian vault. It replaces basic search with a more intuitive, "just works" experience that ranks results by relevance rather than just keyword matching.
How it works
It utilizes the MiniSearch library and the BM25 algorithm to score documents. Relevance is determined by how often query terms appear in the document body, filename, and headings. It can also index PDFs, images, and Office documents when paired with the Text Extractor plugin.
Who it’s for
Obsidian users who need a high-performance, keyboard-centric way to find notes and files, and those who want to expose their vault search to external tools via a local HTTP server.
Highlights
- Smart Ranking: Uses BM25 for automatic document scoring based on content and metadata.
- Broad File Support: Indexes notes, PDFs, images, and Office documents.
- Keyboard-First Design: Mimics the Quick Switcher workflow and supports Vim navigation keys.
- Advanced Querying: Supports quoted expressions, exclusions, and file-type filtering.
- Extensibility: Includes an opt-in local HTTP server for external queries.
Related
- Project
- Project
- Project
- Project
- Project