How do you organize a digital comic collection?

One file per issue, one folder per series

The folder structure matters because servers and apps scan it to build your library. Kavita's wiki is blunt about what it expects: every series nested in its own folder under the library root, no files sitting loose at the top level, and the same series never split across two adjacent folders (Kavita's scanner guide, read 2026-10-01). Komga works the same way: it creates a Series for each subfolder of a library, at whatever depth, and a Book for each file inside it (Komga's library docs, read 2026-10-01).

A layout that works for both servers looks like this:

Comics/
├── Saga/
│   ├── Saga #001.cbz
│   ├── Saga #002.cbz
│   └── Saga #003.cbz
├── Batman/
│   ├── Batman #001.cbz
│   └── Batman #002.cbz
└── Watchmen/
    └── Watchmen.cbz

You do not need publisher or year folders. One level of series folders is enough, and it is what the two big comic servers both document.

Put the details inside the file: ComicInfo.xml

Folder names tell a server what series a file belongs to. ComicInfo.xml tells it the rest: the issue number, the title, the writer, the penciller, the genres, the publisher, the release date. It is a small XML file that sits at the root of the CBZ or CBR archive. The format came from ComicRack, which the Anansi Project says is no longer developed; the project now documents the schema and where it should go next (Anansi Project on ComicInfo.xml, read 2026-10-01).

Komga reads ComicInfo.xml for the release date, writer, penciller, inker, colorist, letterer, cover artist, editor, translator, title, summary, issue number, tags, genres, publisher and more ( Komga's metadata docs, read 2026-10-01). Kavita reads metadata from inside an archive's ComicInfo.xml, and a ComicInfo.xml at the root takes precedence over what Kavita parsed from the file name (Kavita's metadata guide, read 2026-10-01). YACReaderLibrary lists ComicInfo.xml among the ways it organizes a collection. Easemic reads it too: title, writer, penciller and genres come from it into your library.

The reason to tag the files, not just the folders, is that metadata travels with the archive. Move or import the CBZ somewhere new, and the details move with it.

Tagging with ComicTagger

ComicTagger is the tool the engines already point to, and its own repository says what it does: a multi-platform app for writing metadata to digital comics, with automatic issue matching using image processing, batch processing for hundreds of comics at a time, and support for ComicRack and ComicBookLover tagging formats (ComicTagger on GitHub, read 2026-10-01). It fetches comic information from Comic Vine, compares cover images to find the right match, and can rename files from the metadata it writes.

A few practical notes from its user guide. It is happiest when the filename gives it a series name, an issue number (with a # in front) and a year. It can auto-identify a single file by comparing its cover against Comic Vine, or auto-tag a whole batch using whatever metadata is already in the files. Reading and writing RAR and CBR archives takes installing rarlab's command-line tools, rar and unrar. And its wiki warns to back up the CBZs before bulk tagging and work in batches, because automatic matching is not perfect.

The workflow is: normalize the folders first, run ComicTagger over the collection, spot-check the matches it is unsure about, then let it write ComicInfo.xml into every CBZ. After that, the files carry their own catalog.

Convert what does not fit

CBR, CB7 and PDF do not fit every app. CBR is a RAR archive, and rarlab's licence calls the RAR compression algorithm proprietary; CB7 is a 7z. Converting either to CBZ copies the page images into a ZIP, which is an open, documented format. What is a CBZ file, and how do you read one? covers the conversion steps and why CBZ is the format to keep.

Easemic does not open CBR, CB7 or PDF. It imports CBZ and plain ZIP archives and folders of images from the phone. If your collection has CBRs, convert them before you tag them, so ComicTagger and your reader both work on the same files.

Where the files live, and how you read them

After the files are tagged, keep them backed up wherever you keep files: a computer, a NAS, a drive. Then pick how you will read them, because that decides what happens next.

Reading on a phone, no server. Copy the tagged files to the phone and import them into a library app. Easemic (a free Android app, in invite-only testing) imports CBZ and plain ZIP archives and folders of images from the phone, and reads ComicInfo.xml for title, series, writer, penciller, genres and tags. Imported comics are your own files: they read offline, they are never uploaded to storage, and the export leaves them out. Your tagged originals, wherever you keep them, remain the master copy and the backup; on a new phone you copy the folder over and import it again.

Reading with a server. The two self-hosted comic servers the engines already recommend are Kavita and Komga. Kavita serves manga, webtoons, comics and ebooks; Komga serves comic archives, EPUB and PDF. Both scan your folders and build the library from your files. Easemic reads the server's OPDS catalog, never the server's folders, and comics you collect from that catalog go into the storage you connect (S3-compatible storage, WebDAV, OneDrive or Google Drive), so a new phone gets them back by connecting the same storage, with the same password if the library is encrypted. Kavita vs Komga: which comic server should you run? compares the two from their own documentation, and What is S3 storage, and do you need AWS for it? and What is WebDAV, and which drives support it? cover the two main kinds of storage.

The engines today jump from "clean your files" straight to "run a server". The missing middle step is that a tagged, well-foldered collection also works on a phone: copy the files across and import them into an app that reads ComicInfo.xml.

What about Calibre and ComicRack?

Calibre's own user manual calls it an e-book library manager that can view, convert and catalog e-books in most major formats, and its FAQ lists CBZ, CBR and CB7 among its input formats (Calibre's format list, read 2026-10-01). We have not tested Calibre for comics ourselves, so we will not say more than its own pages do.

ComicRack is no longer developed, according to the Anansi Project. Its legacy is ComicInfo.xml, which Komga, Kavita, YACReaderLibrary and Easemic all read. If you are looking for a ComicRack replacement, the honest answer is that there is no single app that does everything it did; the current tools split the job between ComicTagger for tagging, a server for browsing, and a phone app for reading.

Questions people ask