--- name: al-461-get-a-smarter-maintenance-desk-from-your-own-history-have description: "On Oct 22, 2026 Rentvine, a property management software company for single-family rentals, announced Rentvine Cru, a collection of in-app AI agents, at NARPM National. Its first agent focuses on maintenance: it reviews work orders as they come in, suggests a ranked vendor for a new work order, flags a closed issue that resurfaces, catches duplicate requests and drafts routine resident and vendor messages. Its suggestions come from the company's own history in the platform, and a feature called Guidance lets the property manager set rules in plain English. Rentvine also made its MCP connection generally available, with more than 90 tools so managers can query live data from Claude, ChatGPT or other AI tools; one customer quoted in the release said Claude calculated termination fees for a multi-unit owner in minutes. The prompt below does the same review using an export of your past work orders. Use when the user wants help with property maintenance work order review (operations) for…" --- # Get a smarter maintenance desk from your own history: have an AI review each new work order, suggest the best vendor from your past jobs, flag repeat problems and duplicate requests, and draft the messages On Oct 22, 2026 Rentvine, a property management software company for single-family rentals, announced Rentvine Cru, a collection of in-app AI agents, at NARPM National. Its first agent focuses on maintenance: it reviews work orders as they come in, suggests a ranked vendor for a new work order, flags a closed issue that resurfaces, catches duplicate requests and drafts routine resident and vendor messages. Its suggestions come from the company's own history in the platform, and a feature called Guidance lets the property manager set rules in plain English. Rentvine also made its MCP connection generally available, with more than 90 tools so managers can query live data from Claude, ChatGPT or other AI tools; one customer quoted in the release said Claude calculated termination fees for a multi-unit owner in minutes. The prompt below does the same review using an export of your past work orders. ## Inputs to collect first Ask the user for anything below that they haven't already given you. Don't guess missing facts. - E.g. preferred vendors per trade, approval limit $[amount - Amount - CHECK - Your own material (text, notes or data) ## Instructions Do the task below with the user's inputs in place of each [bracket]. Act as my maintenance coordinator. My rules: [e.g. preferred vendors per trade, approval limit $[amount], emergencies = no heat, no water, active leak, gas smell, lockout; owner approval needed above $[amount]]. Past work orders: [attach an export from the last 12-24 months with property, unit, date, category, description, vendor, cost, days to complete, callbacks/notes]. New requests: [paste]. For each new request: 1) Category and urgency under my rules, with one line why. 2) Duplicates: is this the same problem as another open request (same unit or same building system)? 3) Repeat problem: has this unit or building had the same issue closed in the last [6] months? If so, show the earlier work orders and suggest asking for a root-cause fix instead of another patch. 4) Vendor: rank the top 3 vendors for this trade from my own history, using jobs done, average cost, days to complete and callbacks, and show the numbers. 5) Draft a short message to the resident (what happens next, any access or safety notes) and a work order message to the top vendor. Don't dispatch, approve or send anything; I'll confirm each one. Mark anything uncertain [CHECK]. ## 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. - Remove names, contact details, account numbers and other personal data the task doesn't need before you paste, and check your AI tool's data and training settings. - Read and edit the draft before it reaches a customer or goes public. If you later automate this prompt, start with "draft and notify me", not "send automatically". - Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed. ## About this skill - Works with: Claude, ChatGPT, Rentvine, AppFolio, Buildium, DoorLoop, Google Sheets - Based on this real-world example: https://www.prnewswire.com/news-releases/rentvine-expands-ai-leadership-with-rentvine-crus-growing-agent-collection-and-mcp-general-availability-302917235.html - Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-461-get-a-smarter-maintenance-desk-from-your-own-history-have-an/ - 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/