Sort property maintenance photos by issue type
Tenants and staff send photos with maintenance requests, and a decision model sorts each by trade and urgency level so the property manager sees leaks and electrical issues first and can dispatch the right contractor.
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
Small property managers get maintenance requests by text, email, and portal, often with a photo and a vague description. Jev reads the photo and the message together and makes two decisions: which trade the issue belongs to, and how urgent it looks on a fixed scale. Requests where both answers clear a 0.8 confidence are tagged and placed in the right queue. Anything below that, or anything Jev rates as urgent, goes straight to the property manager. Jev does not decide whether a repair is approved, who pays, or anything about the tenancy. The property manager still reviews each request, confirms the urgency with the tenant when needed, and chooses and books the contractor.
Example workflow
- Maintenance requests with photos arrive through your property portal, a shared Gmail inbox, or a web form.
- Make sends the photo and the tenant's message to Jev with the image preview header.
- Jev answers a Choice question, which trade fits this issue, with the allowed answers plumbing, electrical, heating and cooling, appliance, pest, structural, and general, and a Score question rating urgency as routine, soon, or urgent.
- Code tags the request in a Google Sheet or your property software with the trade, urgency, and confidence scores.
- Urgent or low-confidence requests are sent to the property manager in Slack or by text right away.
- The property manager confirms the details, picks the contractor, and replies to the tenant.
Tools
- Jev (TypeSafe AI)
- Gmail
- Make
- Google Sheets
- Slack
- Twilio
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.
- A photo can make a serious problem look minor, such as a small stain from a hidden leak — guardrail: any mention of water, gas, smoke, or sparks in the message is escalated regardless of the photo answer.
- Urgency ratings could delay a real emergency — guardrail: tenants are always told to call the emergency line for safety issues, and the sort never replaces it.
- Requests could be read as tenancy decisions — guardrail: Jev only sorts by trade and urgency; approvals, charges, and tenant matters stay with the property manager.
Related recipe
Support Ticket and Review Triage with Policy-Cited Draft Replies: An agent tags every helpdesk ticket and public review by intent, urgency, and sentiment, routes it, and drafts a reply that cites your help center. People approve refunds and all public replies.
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:
- Gmail (official page): developers.google.com/workspace/gmail (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)
- Twilio MCP server (official page): www.twilio.com/docs/mcp (checked 2026-11-09)