Data sources

Where the plant names, descriptions, and photos in Twigged come from, and how we credit them.

Twigged combines open and reference datasets with hand-curated and model-assisted botanical summaries. Plant fields retain source provenance where it is available. Third-party photographs carry the photographer credit and licence supplied by their source, while original illustrations are identified separately. Below are the main projects and methods used to build the catalogue.

World Checklist of Vascular Plants (WCVP)

Scientific names, family and genus classification, lifeform, and native-range information come from the World Checklist of Vascular Plants maintained by the Royal Botanic Gardens, Kew. WCVP is published under the Creative Commons Attribution 4.0 International licence (CC BY 4.0).

Credit: “The World Checklist of Vascular Plants”, Royal Botanic Gardens, Kew.

USDA PLANTS Database

Growth habit, native status for North America, and taxonomy for US species come from the USDA PLANTS Database, maintained by the United States Department of Agriculture, Natural Resources Conservation Service. As a US federal government work, the data is in the public domain.

Wikidata and Wikipedia

Common names and some descriptions for well-known species are drawn from Wikidata (CC0) and the English Wikipedia (CC BY-SA 4.0). Where a description originates on Wikipedia, the plant page links back to the article for full attribution and edit history.

Royal Horticultural Society references

Where available, Twigged records RHS hardiness bands, Award of Garden Merit references, and Plants for Pollinators classifications, with field-level provenance. The underlying reference material remains the property of the Royal Horticultural Society. Inclusion does not imply RHS endorsement of Twigged.

Horticultural profiles

Sunlight, watering, mature size, flowering period, and care guidance are assembled from cited open sources, hand-curated records, and reviewed model-assisted synthesis based on a plant's botanical identity and independently held reference fields. Model-assisted content is labelled in the field provenance and is guidance rather than a source authority. When independently supported safety or hardiness evidence is unavailable, Twigged leaves those fields unconfirmed instead of inferring them from generated prose.

Plant photographs

Twigged uses a licence-aware resolver to find exact-taxon photographs from iNaturalist, Wikimedia Commons, and GBIF occurrence media. We only accept photos on licences we can use in a commercial product — Creative Commons CC0, CC BY, CC BY-SA, and public domain. When no suitable exact image exists, the catalogue may use a clearly identified representative image of a related plant or an original AI-generated botanical illustration. Third-party photos are stored with their photographer name and licence, and those details appear under the photo on the plant detail page.

If you are a photographer whose photo appears in Twigged and you believe the attribution or licence is wrong, please email hello@twigged.design and we will correct or remove it.

AI-generated content

AI design suggestions, model-assisted catalogue summaries, and visualiser images are generated by models from OpenAI and Anthropic, and, where those optional routes are configured, by Google for image fallback or fal.ai for the depth-conditioned structure stage. These outputs do not have a single human source; their accuracy should always be checked before being used with a client. See our Terms for details.