aifs

Local semantic search. Stupidly simple.

open-sourcememory-knowledge
Visit WebsiteView on GitHub
452
Stars
+38
Stars/month
0
Releases (6m)

Overview

aifs (AI Filesystem) is a local semantic search tool that enables fast, intelligent searching across folders containing mixed file types. It automatically chunks and embeds documents using state-of-the-art parsing and embedding models, storing the results in a local `_.aifs` index file for lightning-fast subsequent searches. The tool supports a wide range of file formats including text files, code, documents, images, and PDFs. Built on top of Unstructured for parsing and Chroma for embedding, aifs intelligently updates its index when files are modified or added, making it ideal for searching dynamic folders like codebases or document repositories. The tool was specifically designed to enable semantic search capabilities for AI tools like open-interpreter, but serves as a general-purpose solution for anyone needing semantic search over local file collections. Its minimalist approach focuses on being the best local semantic search solution rather than expanding into broader functionality.

Pros

  • + Extremely fast searches after initial indexing due to local embedding storage
  • + Supports comprehensive file format coverage including code, documents, images and PDFs
  • + Intelligent incremental updates - only re-indexes changed or new files

Cons

  • - Large dependency footprint when installing full document parsing support
  • - Does not yet handle file deletions from the index
  • - Initial indexing can be time-consuming for large folders

Use Cases

Getting Started

Install with `pip install aifs` and optionally `pip install "unstructured[all-docs]"` for full document type support. Import the search function with `from aifs import search`. Start searching with `search("your query", path="/path/to/folder")` or use current directory by default.