Bookshelf launches as database-less, serverless eBook library

Developer murerkinn has released Bookshelf, a self-hosted eBook library designed to operate without a traditional database. The tool allows users to list, filter, download, and read EPUB and PDF files directly within a web browser.

Bookshelf offers two primary deployment paths: as a Cloudflare Worker utilizing R2 object storage or as a Node.js server running from a local directory. Rather than relying on a live database to track assets, the application uses a dedicated sync tool to build the library and publish it to the user's chosen storage provider. To facilitate reading, the platform includes built-in browser readers for both PDF and EPUB formats. While the system supports user profiles to track reading positions, this data is stored within the library data itself. Notably, Bookshelf does not include built-in authentication; the author recommends deploying the service behind an external authentication layer or within a trusted network.

The shift toward minimal infrastructure

The project arrives at a time when self-hosting enthusiasts are increasingly seeking "lean" alternatives to heavyweight software. For years, the gold standard for personal libraries has been tools like Calibre-web and Audiobookshelf. While powerful, these often require a full virtual private server (VPS) and the ongoing maintenance of database backups and state management. Bookshelf differentiates itself by targeting users who already own their collections and want a lightweight way to serve them without the infrastructure baggage of a relational database.

Lowering the barrier to entry

By removing the database requirement and leveraging object storage, Bookshelf significantly lowers the technical barrier for deploying a personal digital library. The ability to run as a Cloudflare Worker effectively makes the library "serverless," allowing users to maintain a permanent digital shelf with almost zero maintenance. This architecture eliminates the risk of database corruption and simplifies the backup process, as the library's state is tied directly to the object storage.

Future considerations

As the project grows, users will need to weigh the benefits of this minimal footprint against the lack of native security features. Because the author explicitly advises using an external auth layer, the current deployment workflow requires users to be comfortable configuring their own network security or proxy services. Future updates may clarify how the sync tool handles very large libraries or if additional storage providers beyond R2 and local directories will be supported.