scrift

Python SDK for the Scrift brand asset API. Fetch brand colours, SVG logos, raster exports, and metadata for thousands of companies - one API call at a time.

Installation

pip install scrift

Quick start

from scrift import Scrift

with Scrift(api_key="scrf_...") as client:
    # Look up a single brand
    brand = client.catalog.get("stripe")
    print(brand.name, brand.brand_color)  # Stripe 635BFF
    print(brand.rate_limit)  # RateLimitInfo from X-RateLimit-* headers (or None)

    # List the catalog
    page = client.catalog.list(limit=10)
    for item in page.items:
        print(item.slug)

    # Batch lookup (up to 50 slugs)
    batch = client.catalog.batch(["stripe", "github", "vercel"])
    print(batch.found, batch.not_found)

    # Search by name
    results = client.catalog.search("stripe")
    for match in results.matches:
        print(match.slug, match.name)

    # Fetch an SVG logo
    svg_bytes = client.svg.get("stripe", variant="dark")

    # PNG / WebP (size is an API-defined enum, e.g. 32 … 512)
    png_bytes = client.raster.get_png("stripe", size=128, variant="mono")
    webp_bytes = client.raster.get_webp("stripe", size=256)

    # Look up a brand by domain
    info = client.brand.get("stripe.com")
    print(info.variants)  # ['color', 'mono', 'wordmark']

Method reference

Method	Description
client.catalog.get(slug)	Single brand by slug
client.catalog.list(limit=, offset=)	Paginated catalog list
client.catalog.batch(slugs)	Batch lookup (max 50)
client.catalog.search(query, limit=)	Search by name (min 2 chars)
client.svg.get(slug, variant=)	Raw SVG bytes
client.raster.get_png(slug, size=, variant=)	Raw PNG bytes
client.raster.get_webp(slug, size=, variant=)	Raw WebP bytes
client.brand.get(domain)	Brand lookup by domain

SVG variants

mono, color, dark, light, wordmark, icon

Error handling

from scrift import Scrift, NotFoundError, RateLimitError

with Scrift(api_key="scrf_...") as client:
    try:
        brand = client.catalog.get("nonexistent")
    except NotFoundError as e:
        print(e.error_code)  # service_not_found
    except RateLimitError as e:
        print(f"Retry after {e.retry_after}s")

All exceptions inherit from ScriftError and carry status_code, error_code, and message.

Exception	HTTP status
AuthenticationError	401
NotFoundError	404
ValidationError	422
RateLimitError (ScriftRateLimitError)	429 (auto-retries once, capped at 30s)
APIError	everything else

Links

• Scrift
• API Documentation

License

MIT