Keyboard shortcuts

Press ← or → to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

The Catalog System

How Mimir manages D&D 5e reference data and why the catalog works the way it does.

Two Kinds of Data

Mimir’s database holds two fundamentally different kinds of content:

Catalog data is reference material — monsters, spells, items, races, classes, backgrounds, feats, conditions, and more. This data is read-only within Mimir. You search it, browse it, and reference it, but you don’t edit it directly.

Campaign data is your content — campaigns, modules, characters, documents, maps, homebrew. This is fully editable and belongs to you.

The catalog exists to support the campaign. When you add a monster to a module, you’re creating a reference from your campaign data to the catalog. When you add a spell to a character, same thing. The catalog is the shared dictionary; your campaign is the story you write with it.

Import, Not Distribution

Mimir doesn’t ship with D&D content — it can’t, for licensing reasons. Instead, it provides an import system that reads structured data archives and populates the catalog tables.

The import format follows the schema established by the 5etools community project. If you have a compatible data archive, Mimir can import it. The importer handles:

  • Monsters with full stat blocks, traits, and actions
  • Spells with casting details and class lists
  • Items including weapons, armor, and magic items with variant generation
  • Races, classes, subclasses, backgrounds, feats
  • Conditions, languages, deities, and many other entity types
  • Token images and creature artwork
  • Source book metadata and publication information

After import, all data lives in your local SQLite database. Mimir never phones home or downloads content.

Source Filtering

Not every campaign uses every sourcebook. A “Player’s Handbook only” game shouldn’t show Xanathar’s Guide spells in character creation.

Mimir’s source filtering works at two levels:

Campaign sources control which books are active for a campaign. When you open Campaign Sources and enable only PHB and MM, catalog searches within that campaign only return content from those books.

Character sources optionally restrict further. A character can have their own source list — useful when one player owns different books than another.

This filtering is consistent everywhere: spell searches, monster searches, item searches, and character creation all respect the active source configuration.

Full-Text Search

The catalog uses SQLite’s FTS5 (Full-Text Search) engine. When you search for “fire,” it matches:

  • Monster names: “Fire Elemental”, “Fire Giant”
  • Spell names: “Fireball”, “Fire Bolt”
  • Item descriptions: a sword with fire damage
  • Monster abilities that mention fire resistance

The FTS index covers both rule text and flavor text, using the Porter stemming algorithm so “casting” matches “cast” and “dragons” matches “dragon.”

Homebrew Integration

Homebrew content (custom items, monsters, spells) lives alongside catalog data in search results. When you search for monsters to add to a module, homebrew monsters from the active campaign appear in the same results list, marked with an HB badge.

This blending is intentional — you shouldn’t have to remember whether that custom dragon variant is homebrew or catalog. It just shows up where you need it.

See The Homebrew System for details on creating custom content.

Data Structure

Catalog entities are stored as structured JSON in the database, following the schema conventions of the source data format. This means:

  • Monster stat blocks include everything: abilities, actions, legendary actions, lair actions, regional effects
  • Spells include casting components, duration, scaling behavior
  • Items include variant generation rules for magic items (a +1/+2/+3 weapon generates three entries)

The JSON is parsed on-demand by the frontend formatters, which render stat blocks, spell descriptions, and item cards in familiar D&D presentation style.

Why Not a Bundled Database?

You might wonder why Mimir doesn’t ship with a pre-populated catalog. Three reasons:

  1. Licensing. D&D content is owned by Wizards of the Coast. Distributing it would require a license Mimir doesn’t have.
  2. Freshness. The community data format is actively maintained and updated. Importing lets you use the latest corrections and additions.
  3. Choice. You control exactly which sources are in your database. No bloat from content you don’t use.

See Also