Guide sections

PDF adventure import

Extract proposed creations from adventure PDFs, review the queue, and approve only the material you want to keep.

This is the Adventure path. If you want to import a rulebook so the engine follows your game’s mechanics instead, see Game system rules - same wizard, different output.

Before you start

Before you import: the text of this PDF leaves your machine and goes to Anthropic so it can be read. Importing does not keep the book local. The file itself never moves, but the text extracted from it does. Check you’re comfortable with that for this particular book, and see Privacy for what happens to it there.

PDF extraction is an AI feature, so it needs an AI subscription (Companion or Power). It reads the document, identifies content, and writes structured creations for you - that’s the engine doing the work.

You’ll also want a text-based PDF. Scanned-image PDFs with no selectable text have little for the engine to read.

How it works

You pick a PDF and the pages you care about. The engine converts those pages to text, extracts the content it recognizes, and hands you a review queue. Nothing is added to your campaign until you approve it.

What Gets ExtractedBecomes
Named charactersNPC creations
Places and roomsLocation creations
Treasure and gearItem creations
Set-piece fightsEncounter creations

Every extracted creation keeps a page reference back to where it came from in the PDF.

Step by step

1. open the import wizard

  1. Open the Library from the sidebar and click Import to Library
  2. Click Import from PDF
  3. Choose your PDF file
  4. On the category screen, choose Adventure (not Sourcebook)

The PDF adventure import is a 9-step wizard - analysis, page selection, extraction, image selection, creation review, error/duplicate resolution, confirm, and complete - walked through below.

2. analyze and select pages

The wizard analyzes the document and shows a page-by-page preview. Select the pages you want to import - or keep the default selection of all pages. Skipping front matter, ads, and indexes keeps extraction focused and faster.

3. let the engine extract

The engine processes your selected pages and pulls out the content it recognizes. Larger page ranges take longer; you’ll see progress as it works.

4. choose images (optional)

If the pages contain artwork, you can pick which images to bring in and attach to your creations.

5. review the queue

This is the important step. Every extracted creation appears in a review queue, and nothing imports until you approve it. Approved creations import to the Master Catalog (your Library) as-is, to preserve the source’s integrity - distinct from your current campaign; you bring them into a campaign separately afterward. For each one you can:

  • Approve it as-is
  • Change its type if the engine guessed wrong (for example, a location that should be an NPC)
  • Delete it if it shouldn’t be imported at all

You can’t edit fields here - the queue is for approving, retyping, or removing. Approve a creation first, then refine it like any other.

There’s a batch approve to accept everything visible at once when the extraction looks good.

6. resolve errors and duplicates

The wizard flags likely problems - duplicates of creations you already have, low-confidence extractions, or inconsistent details - so you can fix or skip them before finishing.

7. confirm and finish

Review the summary of what will be imported, confirm, and Campaign Familiar adds your approved creations to the Master Catalog / Library as-is. The completion screen shows what was created, anything skipped, and any warnings.

After import

  • Spot-check page references. Open a few creations and confirm the details match the source pages.
  • Connect the dots. Add relationships between the imported NPCs, locations, and factions, or let the engine suggest connections.
  • Tag the batch. A tag like imported-pdf makes it easy to find and tidy everything later.

Tips

  • Adventure vs Sourcebook. Choose Adventure to extract NPCs, locations, and encounters as creations. Choose Sourcebook to extract tables, rules, and reference material for the engine to follow - that’s the Game system rules path.
  • Trim the page range. Importing only the chapters you need is faster and produces a cleaner review queue.
  • Review, don’t rubber-stamp. AI extraction is a strong first pass, not a final answer - the review queue is where you make it yours.

Troubleshooting

Extraction found very little

  • The PDF may be a scanned image with no selectable text, or you may have selected pages with little content. Try a text-based PDF or a richer page range.

The engine mislabeled some creations

  • Use Change type in the review queue to fix it before approving (you can edit the creation after import).

Duplicates of creations i already have

  • The wizard’s error-resolution step flags likely duplicates so you can skip them or merge by hand after import.

”needs a subscription”

  • PDF extraction is an AI feature and needs a Companion or Power subscription. See Subscription tiers.