PDF Search for Zed logo

PDF Search for Zed

by freespirit

A document search extension for Zed that enables semantic search through PDF documents, allowing users to leverage the results within Zed's AI Assistant. It enhances the AI Assistant's context by providing relevant sections from the PDF.

View on GitHub

Last updated: N/A

What is PDF Search for Zed?

PDF Search for Zed is a Zed extension that allows you to semantically search through PDF documents and integrate the results into Zed's AI Assistant. It uses OpenAI's API to generate embeddings and find relevant sections based on your query.

How to use PDF Search for Zed?

  1. Clone the repository. 2. Set up the Python environment using uv. 3. Install the Dev Extension in Zed. 4. Build the search database using the provided script and your OpenAI API key. 5. Configure Zed to use the context server. 6. Open Zed's AI Assistant and type /pdfsearch followed by your query.

Key features of PDF Search for Zed

  • Semantic search through PDF documents

  • Integration with Zed's AI Assistant

  • Uses OpenAI embeddings (with plans for self-contained alternative)

  • Supports multiple PDFs

  • Supports additional file formats (optional)

Use cases of PDF Search for Zed

  • Quickly find relevant information within large PDF documents

  • Enhance the context of Zed's AI Assistant with specific document sections

  • Research and analysis using AI-powered search

  • Code documentation search

FAQ from PDF Search for Zed

What is required to use this extension?

Currently, you need an OpenAI API key and uv installed on your system.

How do I build the search database?

You need to run the rag.py build script with the path to your PDF files or directories, providing your OpenAI API key.

Can I use this with other file types besides PDF?

The extension has optional support for additional file formats beyond PDF.

Is there a way to use this without an OpenAI API key?

The developers plan to implement a self-contained alternative for embeddings in future versions.

Does it support multiple PDFs?

Yes, it supports multiple PDFs. You can provide multiple files and directories as arguments to the rag.py build script.