Source images and scans with links
Connect
SPIRITT
Google Cloud VisionImages to insights
Ready for real work
SPIRITT turns Google Cloud Vision annotations into searchable records, review queues, and document intake systems. It also builds and maintains Product Search catalogs with products, product sets, and reference images.
What you can do with Google Cloud Vision.
Documents into usable records
Extract text from images and supported files, preserve source links, and route uncertain fields into a review queue instead of treating every result as fact.
Visual signals into action
Use image labels, landmark detections, and explicit content tags to organize assets and prioritize review against your own rules.
A maintained product library
Create products and reference images, organize product sets, and keep names, descriptions, and product labels aligned with your catalog.
Works with your other apps.
Get started in three steps.
01
Open your workspace
Start a SPIRITT workspace for your document intake, image review, or product reference library operation.
02
Connect Google Cloud Vision
Connect Google Cloud Vision with an API key and appropriate access to the project and image or file resources your mission needs.
03
Delegate an outcome
Give SPIRITT your source files, catalog, and processing rules. It builds the workflow, runs annotations, and carries results into your team's systems.
Hand over a mission.
Choose a mission. Make it yours.
Put Google Cloud Vision to work.
Turn visual inputs into finished work
Questions
Google Cloud Vision, with SPIRITT.
01What can SPIRITT extract from images and documents?+
Google Cloud Vision supports image labeling, OCR, face and landmark detection, and explicit content tagging. File annotation supports PDF, TIFF, and GIF inputs. SPIRITT can turn returned text and annotations into records and workflows; mapping OCR text into business fields is additional processing, not a guarantee that every field was read correctly.
02What access does this integration need?+
The connection uses an API key. The mission also needs appropriate access to the Google Cloud project and the image or file resources being processed. Product set imports use a CSV stored in Google Cloud Storage, so that source must be accessible for the import.
03Can SPIRITT keep large annotation jobs moving?+
Yes. It can submit asynchronous image and file batches and retrieve operation status while processing continues. SPIRITT can maintain a ledger of submitted inputs, completed outputs, and failures, then route exceptions and coordinate retries without losing track of successful work.
04Where can the annotation results go?+
SPIRITT can connect OCR results to Google Sheets records, organize image annotations in Airtable, store a processing history in Supabase, and coordinate reviews through Slack or Asana. Those connected systems hold the business records and human decisions, while Google Cloud Vision supplies the visual analysis.
05What can SPIRITT build around these capabilities?+
Useful builds include a scanned-document intake desk, an image moderation queue, an equipment-label capture app, and a product reference library admin panel. Each can retain source images or links, expose uncertain results for review, and carry accepted records into the team's operational tools.
06How is product library management different from image labeling?+
Image labeling returns annotations about supplied images. Product Search library management uses persistent Products, ProductSets, and ReferenceImages. SPIRITT can create and update product records, manage set membership, import libraries, and maintain reference images. A maintained library is distinct from a finished shopper-facing visual search experience.