--- name: al-485-decide-how-your-business-answers-when-an-ai-agent-calls-w description: "More and more phone calls to local businesses come from AI agents asking questions or trying to book on someone's behalf, and staff often don't know whether to answer or hang up. On October 29, 2026 SF Times summarised SF Standard reporting on test calls by the startup Voygr: its AI agents placed 848 calls to San Francisco small businesses with simple questions such as the price of a salon's cheapest haircut. About 1 in 3 stayed on the line after hearing it was an AI agent, 1 in 25 already had their own AI agent answering, vets and dentists were most likely to keep talking and dry cleaners and bars most likely to hang up. A restaurant group said agents often got stuck at payment, and a bike shop owner wants them to say who built them. The figures come from a company testing its own software. The prompt below drafts a simple house policy with any AI assistant. Use when the user wants help with phone policy for ai agent callers (customer support) for a small business." --- # Decide how your business answers when an AI agent calls: write a short policy for what your staff or phone bot shares with AI callers (prices, hours, bookings, payments), what needs a human, and what to ask the caller to confirm More and more phone calls to local businesses come from AI agents asking questions or trying to book on someone's behalf, and staff often don't know whether to answer or hang up. On October 29, 2026 SF Times summarised SF Standard reporting on test calls by the startup Voygr: its AI agents placed 848 calls to San Francisco small businesses with simple questions such as the price of a salon's cheapest haircut. About 1 in 3 stayed on the line after hearing it was an AI agent, 1 in 25 already had their own AI agent answering, vets and dentists were most likely to keep talking and dry cleaners and bars most likely to hang up. A restaurant group said agents often got stuck at payment, and a bike shop owner wants them to say who built them. The figures come from a company testing its own software. The prompt below drafts a simple house policy with any AI assistant. ## Inputs to collect first Ask the user for anything below that they haven't already given you. Don't guess missing facts. - Type, e.g. salon, restaurant, dental office, bike shop - Staff, an answering service, our own AI receptionist - Prices, hours, availability, bookings, orders - CHECK ## Instructions Do the task below with the user's inputs in place of each [bracket]. Help me write a short policy for how my business handles phone calls from AI agents acting for customers. Business: [type, e.g. salon, restaurant, dental office, bike shop]. Who answers our phone: [staff, an answering service, our own AI receptionist]. Things callers usually ask: [prices, hours, availability, bookings, orders]. 1) List which questions we can safely answer for an AI caller (public info like hours and starting prices) and which need a human or a callback. 2) Write a short script for staff: how to ask the caller who it is acting for and which company built it, and what to say if it won't say. 3) Set rules for bookings and payments by AI callers: what details we need, whether we need a card or deposit, and how we confirm with the real customer by text or email. 4) Write the same rules as instructions for our own AI receptionist, if we have one. 5) Suggest a simple log so we can count AI calls for a month and decide if it's worth more effort. Mark anything that depends on local law or card-network rules as [CHECK]. Keep it to one page. ## Safety checks (share these with the user) - 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. - 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, Google Docs - Based on this real-world example: https://www.sftimes.com/brief/2026-10-29/6d5f0a-press-1-if-you-re-a-robot-ai-agents-keep-calling-local-businesses/ - Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-485-decide-how-your-business-answers-when-an-ai-agent-calls-writ/ - 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/