CaraArchive Index Dataset

1) This is an Index Dataset, not an Image dataset

• This dataset contains 0 image data.
• It only contains links to Cara App's CDN and metadata.
• HuggingFace may load some images in the preview because it detects the CDN links, however those images do not exist as data in this dataset, HF literally just loads them using your own connection as a preview.

---

How to Query the Database with tool.py (make sure catalog.db is in the same directory)

To see a clean summary overview of an artist:

python tool.py --user artist_fc17f3c7 --overview
# or query by author UUID:
python tool.py --user fc17f3c7-73ab-4866-8d1a-554da8ea68a9 --overview

To see full breakdown and all image links by artist:

python tool.py --user artist_fc17f3c7

To see all artworks, software, metadata, and creator for a specific post:

python tool.py --post 00002b4d-120a-4e94-bae7-49b9c905bc05
# or using web URL:
python tool.py --post https://cara.app/post/00002b4d-120a-4e94-bae7-49b9c905bc05

To find the post and creator that an artwork image belongs to:

python tool.py --art tenaebron-S4TIL677BLm8bh_T0jB6I-post-cover-image.jpeg
# or using direct CDN link:
python tool.py --art https://cdn.cara.app/production/posts/93d6c843-1f4a-429d-8144-31e9099b95c6/tenaebron-S4TIL677BLm8bh_T0jB6I-post-cover-image.jpeg

---

Database Structure (catalog.db)

The SQLite database contains 3 main tables (and 2 tracking tables):

master_artworks_12m (8,521,351 rows)

Contains all full-resolution master artwork uploads.

• id: Unique image ID (<post_id>_slide_<number>) [PRIMARY KEY]
• post_id: Post ID this artwork belongs to
• author_slug: Anonymized creator handle (artist_<hex>)
• slide_number: Order in the multi-image gallery (1, 2, 3...)
• title: Post title (with personal handles/links redacted as [CENSORED])
• cdn_url: Direct high-res CDN link (UNIQUE)
• width: Image width in pixels
• height: Image height in pixels
• created_at: Upload timestamp

artworks (3,431,475 rows)

Contains all posts and cover thumbnails.

• id: Unique image ID (<post_id>_<suffix>) [PRIMARY KEY]
• author_slug: Anonymized creator handle (artist_<hex>)
• author_id: Anonymized creator UUID
• post_id: Post ID this artwork belongs to
• cdn_url: Cover image / thumbnail link (UNIQUE)
• created_at: Creation timestamp
• downloaded: Download status flag (0 = pending, 1 = downloaded)

post_metadata (3,431,474 rows)

Contains all tags, software, descriptions, metrics, and audit statuses.

• post_id: Post ID [PRIMARY KEY]
• author_slug: Anonymized creator handle (artist_<hex>)
• author_name: Anonymized display name (Artist <HEX>)
• title: Post title (with personal handles/links redacted as [CENSORED])
• content: Description / lore text (with social handles, emails, and links redacted as [CENSORED])
• softwares: Software used (JSON array, e.g. ["Blender", "Photoshop"])
• tags: Categories and tags (JSON array, e.g. ["Illustration", "ConceptArt"])
• raw_tags: Typed tags (JSON array of objects, e.g. [{"type": "category", "value": "Horror"}])
• like_count: Number of likes
• comment_count: Number of comments
• status_code: HTTP status (200 = active, 404 = deleted post, 403 = private)
• is_deleted: 1 if post was removed, 0 if active
• created_at: Creation timestamp

Additional Tracking Tables:

• indexed_artists (3,267 rows): author_id (PK), slug, artwork_count
• processed_posts (3,431,475 rows): post_id (PK)

---

Files in this Release

• catalog.db: SQLite database with all tables and B-Tree indexes (9.82 GB)
• tool.py: CLI dataset explorer script
• README.md: Dataset documentation