Check menu and dish photos for allergen labels before posting
Before a menu board or dish photo goes on the website or delivery apps, a decision model checks whether allergen labels are visible on it and holds back any image missing them for the kitchen manager.
Uses image input, which TypeSafe AI labels a preview feature for Jev. Test it on your own sample images before relying on it.
An example pattern written by Agents Lookup. We have not yet found a public build of this exact job.
The use case
Restaurants and cafes post menu boards and dish photos to their website and delivery apps, and many have agreed or are required to show allergen information. Jev looks at each image and answers whether allergen markings are visible, whether a key to the symbols is shown, and whether the text is readable at normal size. Images that clear a 0.9 confidence on all three are marked ready to post. Anything that fails or is unsure goes to the kitchen manager with the reason. Jev does not decide which allergens a dish contains; that comes from your recipes and suppliers. The kitchen manager checks the labels are correct, approves every image before it is posted, and owns the allergen information.
Example workflow
- New menu photos or menu board designs are dropped into a Google Drive folder by whoever updates the menu.
- Make sends each image to Jev with the image preview header.
- Jev answers three Noul questions: allergen symbols or labels are visible on the menu items, a key explaining the symbols is shown, and the menu text is readable at normal viewing size.
- Images that clear the threshold on all three are listed as ready in a Google Sheet; the rest are posted to Slack with the failed check named.
- The kitchen manager checks every ready image against the recipe allergen sheet before approving it.
- Only approved images are posted to the website and delivery apps, by a person.
Tools
- Jev (TypeSafe AI)
- Google Drive
- Make
- Google Sheets
- Slack
- Canva
Risks and guardrails
- Image input is a preview feature in Jev 1.14 and TypeSafe says it is not production-ready yet — guardrail: test it on a few hundred of your own photos against a person's labels before relying on it, and keep a person checking every answer until the results hold up.
- Labels being visible does not mean they are correct — guardrail: the kitchen manager checks every label against the recipe sheet before posting.
- Small symbols on a busy board can be missed — guardrail: set a high threshold and send anything unsure to a person.
- Rules on allergen display vary by place — guardrail: confirm your local requirements and treat this check as a reminder, not a compliance sign-off.
Related recipe
Repurpose One Video or Webinar into Posts, Newsletter, Clips, and a Blog Outline: An agent turns one long video transcript into LinkedIn posts, a newsletter section, timestamped clip suggestions, and a blog outline in your brand voice, with a claims check and human approval before publishing.
Want it set up for you? Request a quote.
Sources
Written by Agents Lookup. Credits and official pages for the tools this use case names:
- Google Drive (official page): developers.google.com/workspace/drive (checked 2026-11-19)
- Make MCP server (official page): developers.make.com/mcp-server (checked 2026-11-09)
- Slack MCP server (official page): docs.slack.dev/ai/slack-mcp-server (checked 2026-11-09)
- Canva Skills (official page): github.com/canva-sdks/canva-claude-skills (checked 2026-11-09)