--- name: al-544-find-the-building-problem-hiding-in-your-repair-requests description: "Elevate AI's case study describes a regional property manager with about 800 units across 12 properties whose maintenance requests arrived by phone, email and text, with roughly 10% getting lost. Once an AI assistant and maintenance workflow logged and categorized every request in one place, the team saw that 3 of the 12 properties produced 60% of all plumbing requests, clustered in the same building wings on the same original copper lines. The operations director had wanted to re-pipe those wings for years but lacked data to justify it. A $22,000 re-piping of the worst wing cut emergency plumbing calls there by 85% the next quarter, removing about $8,000 a month in emergency vendor fees (figures from the vendor's case study). You can run the same check on your own maintenance history even before you automate anything: export the tickets, have AI categorize them and look for repeat clusters. Use when the user wants help with maintenance pattern analysis (operations) for a small…" --- # Find the building problem hiding in your repair requests: have AI sort a year of maintenance tickets by property, building wing and type, map where the same repair keeps coming back, and work out whether one planned fix would cost less than the emergency call-outs Elevate AI's case study describes a regional property manager with about 800 units across 12 properties whose maintenance requests arrived by phone, email and text, with roughly 10% getting lost. Once an AI assistant and maintenance workflow logged and categorized every request in one place, the team saw that 3 of the 12 properties produced 60% of all plumbing requests, clustered in the same building wings on the same original copper lines. The operations director had wanted to re-pipe those wings for years but lacked data to justify it. A $22,000 re-piping of the worst wing cut emergency plumbing calls there by 85% the next quarter, removing about $8,000 a month in emergency vendor fees (figures from the vendor's case study). You can run the same check on your own maintenance history even before you automate anything: export the tickets, have AI categorize them and look for repeat clusters. ## Inputs to collect first Ask the user for anything below that they haven't already given you. Don't guess missing facts. - Export with date, property, unit, building/wing, description… - Your own material (text, notes or data) ## Instructions Do the task below with the user's inputs in place of each [bracket]. Here is my maintenance request history for the last [12] months: [paste or attach export with date, property, unit, building/wing, description, vendor, cost, emergency yes/no]. 1. Categorize each request (plumbing, electrical, HVAC, appliance, roof/leaks, pests, other) and flag ones that look like repeats of the same problem. 2. Build a table of request counts and total cost by property, building or wing, and category. Show the top 5 clusters where the same kind of repair keeps coming back. 3. For each cluster, list the likely shared cause to check (age of pipes, roof section, a specific system), stated as a question for my maintenance lead, not a conclusion. 4. For the top 2 clusters, compare what the repeat and emergency repairs cost per year with a planned fix: I'll give you a quote of [$X] or ask me for one. Show the payback in months. 5. Note any requests that seem to have been lost or never closed. Use only my data; mark any category you're unsure of. Keep it to one page I can take to the owner. ## Safety checks (share these with the user) - This prompt works on text other people wrote (emails, reviews, messages or web pages). That text can hide instructions aimed at AI, such as "ignore your rules and…". Paste it as material to work on, and never let the AI act on requests inside it, like paying, sharing logins or changing records. - Check every figure, price and total against your own records before anything is sent, posted or booked. A person approves refunds, discounts, payments and price changes. - Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details. - Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed. ## About this skill - Works with: ChatGPT, Claude, spreadsheet, property management software export - Based on this real-world example: https://elevate-ai.ai/case-studies/property-management-ai - Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-544-find-the-building-problem-hiding-in-your-repair-requests-hav/ - To install in an agent that supports Agent Skills, save this file as SKILL.md in a folder with the same name as `name` above (for example in your skills directory). Install steps for Claude.ai, Claude Code, Cursor and Codex: https://agentic-library.netlify.app/skills/