Spot commercial tenants who might not renew before it's too late: have an AI agent score each tenant's health from building visits, service requests, survey answers and lease dates, then draft an action for each at-risk tenant for you to approve: AI agent prompt

Help me build a tenant health check for my commercial property so I can act before renewals are at risk. Building: [type, size, number of tenants]. Data I can export: [tenant list with suite, size, rent, lease expiry date; access card or visitor counts by month; service requests with dates and categories; tenant survey scores or comments; late payments; amenity or event attendance]. 1) Propose a 0-100 tenant health score from these signals, show the weights and explain each one in plain words, and let me change them. 2) Score every tenant, then list the at-risk ones: low score, a falling trend over 3 months, or a lease ending within [12] months, with the main reasons for each. 3) For each at-risk tenant, suggest one specific action (e.g. a call from the property manager, fixing a repeated service issue, a space review, an early renewal conversation) and draft a short, friendly message for me to edit. 4) Summarise building-wide patterns, such as one floor or one type of complaint driving problems. Don't contact any tenant or change any system; only suggest, and mark any score based on missing or thin data.