Read Vision annotations and product catalogs, then make changes with approval
Connect Google Cloud Vision and your agent can run label, text, and landmark detection on images, check the state of long-running annotation jobs, and inspect Product Search catalogs. Catalog changes — creating products, updating product sets, importing from Cloud Storage, deleting — run as previewed steps you approve. Luumen authenticates with an API key.
29 tools: 13 read, 16 write. Reads answer instantly. Writes require approval by default. Everything is logged.
What a governed Google Cloud Vision run looks like inside Luumen.
Authorize once with API token. Luumen lists the scopes each action needs before you approve the connection, and credentials never appear in the chat.
Read actions answer immediately. Anything that writes — async batch annotate files, annotate images async batch, create vision product, create product set, and more — is shown as a plan and requires approval by default, including the 6 actions classified as destructive. Administrators configure that per tool, so you decide exactly which actions can ever run unattended.
You decide. Actions are granted per agent, skill, and team, and per environment — production is not staging. Read access can be broad while writes stay narrow.
Every call to Google Cloud Vision — read or write, approved or declined — is recorded with the actor, the input, and the result, and can be linked to the ticket or change record.
Connect in minutes. Every action scoped, approved, and audited from day one.