# Auto-Respond — AI answering service and voice agent that answers calls and marketplace texts, qualifies leads and books jobs. > Source: The Agents Index — https://theagentsindex.com/auto-respond (structured, researched, re-verified) > Facts last verified: 2026-09-11 Auto-Respond is a lead-response platform for home-service businesses whose AI voice agent answers inbound calls, calls new leads back and runs outbound campaigns. The same account also auto-replies to Yelp, Thumbtack, Facebook, Instagram and Google LSA enquiries and files them into a CRM. | Fact | Value | | --- | --- | | Website | https://auto-respond.com/ai-answering-service | | Price | from $10/month | | Free trial | Partly | 7-day trial, offered only to eligible first-time customers and locations and only when the signup or checkout order summary shows it. While it runs, Voice Agent limits are 1 phone number, 10 call minutes and 5 recipients per batch campaign. No standing $0 plan exists. ## What it does **Writes code** · No — Scope is inbound and outbound phone and messaging lead handling; nothing on the site claims code authoring. > Voice Agent is the AI phone system inside Auto-Respond. You configure one or more agents, assign them a phone number, and the agent handles calls on your behalf > — https://auto-respond.com/docs/voice-agent/ · Voice Agent docs, opening paragraph **Debugs and fixes** · No — The product diagnoses nothing in software; it runs a qualifying conversation with a lead. > An AI answering service picks up the phone and the texts the moment a lead reaches out, qualifies them against your real pricing and service area, and books the appointment on your calendar. > — https://auto-respond.com/ai-answering-service/ · AI Answering Service page, hero paragraph **Writes tests** · No — No test generation of any kind; outputs are spoken replies, drafts, bookings and call summaries. > Voice Agent is the AI phone system inside Auto-Respond. You configure one or more agents, assign them a phone number, and the agent handles calls on your behalf > — https://auto-respond.com/docs/voice-agent/ · Voice Agent docs, opening paragraph **Reviews code** · No — The only review loop documented is a human reviewing the AI's draft reply, not the AI reviewing a diff. > Yelp requires staff to review AI-generated replies before sending and requires clear disclosure when AI is used. > — https://auto-respond.com/docs/lead-sources/yelp/ · Yelp lead source docs, intro paragraph **Whole-codebase work** · No — There is no repository concept; the agent's context is uploaded reference documents and prompt fields. > Upload documents the agent references during calls — service lists, FAQs, pricing sheets, policies. > — https://auto-respond.com/docs/voice-agent/agent-configuration/ · Agent Configuration, Tab 2 Knowledge Base, Reference Library **Ships it** · No — Its terminal action is a confirmed calendar booking and a CRM handoff, not a deployment. > When on: the agent can check availability and create bookings during the call using a connected scheduling platform. > — https://auto-respond.com/docs/voice-agent/agent-configuration/ · Agent Configuration, Tab 3 Scheduling, Appointment Scheduling toggle **Done-for-you service** · No — You configure and run the software yourself: the vendor documents a self-serve setup of forwarding your line, loading services and pricing, and linking a calendar, with no build-and-run engagement offered. > You forward your business line to the AI agent, add your services, pricing ranges, service area, and hours, set your qualifying questions, link your calendar, and turn it on. > — https://auto-respond.com/ai-answering-service/ · Frequently asked questions, 'How fast can I set up an AI answering service?' ## How much it does unattended **Runs autonomously** · Partly — On the phone it runs the whole job unattended - qualifies, checks the calendar and books - but on Yelp the AI only drafts a reply for staff to review, and text conversations stop after five or ten AI replies. > Start to finish, the lead can be qualified and booked inside a single call. For Yelp messages, the AI prepares a draft for staff review. Other messaging sources follow their own reply rules. > — https://auto-respond.com/ai-answering-service/ · Section 'How an AI answering service works', closing paragraph **Multi-agent** · No — You can create many agents, but each call is handled by exactly one of them and escalation goes to a human phone number in E.164 format; the full agent reference documents no agent-to-agent handoff or orchestration. > Transfer to Phone Number - required. E.164 format. Must use the same country code as the agent's phone number. > — https://auto-respond.com/docs/voice-agent/agent-configuration/ · Tab 5 - Advanced, Call Transfer field list **Agent permissions** · Partly — It acts unattended inside limits you preset - pricing on or off, services it must decline, transfer criteria, custom rules - but there is no per-action approval queue; a human reply in the thread is what stops it. > A list of special instructions, policies, or behavioral guidelines specific to your business. Each rule has an enable toggle and a label. > — https://auto-respond.com/docs/voice-agent/agent-configuration/ · Tab 5 - Advanced, Custom Rules ## Which models it runs on **Model choice** · No — The vendor picks the model: the agent editor's own complete field reference covers six tabs and contains no model picker, no provider setting and no place to attach inference credentials. > This page is the full reference - every field, every toggle, and what each one does. > — https://auto-respond.com/docs/voice-agent/agent-configuration/ · Introduction, describing the six-tab agent editor _Not established: Claude, GPT, Open models_ ## Where you use it **In your editor** · No — Surfaces are the hosted app, the phone line and the connected marketplace inboxes; no editor plugin exists. > Forward your business line to the AI agent and connect the lead channels you sell on. No new hardware. > — https://auto-respond.com/ai-answering-service/ · Set it up in under 2 minutes, step 2 **On the command line** · No — Setup is described entirely through the web app; no terminal tool is documented. > You forward your business line to the AI agent, add your services, pricing ranges, service area, and hours, set your qualifying questions, link your calendar, and turn it on. > — https://auto-respond.com/ai-answering-service/ · FAQ, How fast can I set up an AI answering service? **In your pipeline** · No — The programmatic surface is an outbound webhook plus a reply API for lead conversations, not pull requests or CI jobs. > Auto-Respond sends an HTTP POST request to an endpoint you control every time a lead event happens on this source — a new lead, new contact info, a new message. > — https://auto-respond.com/docs/integrations/rest-webhook/ · REST Webhook docs, intro paragraph **In a browser** · Yes — Everything is run from a hosted web app - sources dashboard, messaging settings, lead conversations and a Jobs board - which the vendor shows as real product screens. > Real screens from the Auto-Respond product tour. Connect a source, teach the AI your business rules, follow every conversation, and see the appointment land in Jobs. > — https://auto-respond.com/ · Section 'INSIDE THE APP', intro line above the screenshot carousel ## Whose machine it runs on **Self-hosted** · No — The vendor describes its product as hosted business software and the pricing page offers only per-source subscriptions on its own cloud; no self-hosted or bring-your-own-infrastructure option appears on any tier. > collects, uses, discloses, and retains Personal Data in connection with our websites, hosted business software, communications, support, and related services > — https://auto-respond.com/privacy/ · Privacy Policy, opening paragraph defining the Services **Open source** · No — Closed and proprietary: the Terms grant only a right to access the hosted Services and expressly forbid attempting to discover the source code or the nonpublic models behind it. > reverse engineer, decompile, disassemble, or attempt to discover source code or nonpublic models, except where law prohibits this restriction > — https://auto-respond.com/tos/ · Section 12, NZ Leads intellectual property, 'Customer may not' list ## What it costs to run **How it meters** · Yes — Two meters, both published: a flat monthly price per connected lead source or phone number, plus voice billed per talk minute, so cost tracks channels activated and minutes spoken rather than seats. > Auto-Respond is priced per source, per month. A source is one connected account or profile on a platform - a single Yelp location, one Thumbtack Pro account, one Facebook Page, and so on. > — https://auto-respond.com/docs/pricing/ · Pricing, opening paragraph above the source pricing table **Free tier** · Partly — There is a 7-day trial and no sales call is required, but it is conditional on being a first-time customer and location and shown in the order summary; there is no standing free plan. > A 7-day trial applies only to eligible first-time customers and locations when the exact signup or checkout order summary shows it. > — https://auto-respond.com/docs/pricing/ · Trial eligibility, first bullet **API access** · Partly — You can drive conversations from your own stack through a per-source Lead Reply API plus an outbound REST webhook, but the vendor publishes no endpoint reference, base URL or general-purpose API for agents, calls or reporting. > The Lead Reply API lets an external system send replies back into Auto-Respond so they're delivered to the lead as if your team sent them from inside the app. > — https://auto-respond.com/docs/integrations/lead-reply-api/ · Lead Reply API, opening paragraph **MCP server** · No — The docs enumerate every integration across three sub-tabs and the closest category lists only Call Bridge, Voice Agent, REST Webhook and Lead Reply API; no MCP server appears anywhere in the documentation. > Other - integrations that do something more specialized than a simple alert or a CRM push. This is where Call Bridge, Voice Agent, REST Webhook, and Lead Reply API live. > — https://auto-respond.com/docs/integrations/ · Section 'How the Integrations tab is organized', third bullet **Bring your own key** · No — Inference and telephony are the vendor's to supply and meter: the complete agent-editor reference exposes no model, provider or API-key field, voice is sold per billable minute, and the vendor states you do not bring your own carrier. > You don't need to bring your own phone service - Auto-Respond sends the text for you. > — https://auto-respond.com/docs/integrations/sms/ · SMS integration, opening paragraph ## Buying it for a team **A company can buy it** · Yes — Sold to businesses, not individuals: sources are owned by a business entity, multi-location and multi-brand organisations get separate businesses, and invoicing runs per legal entity. > For a multi-location or multi-brand organization, each business can hold its own set of sources so messaging, branding, and reporting stay separate. > — https://auto-respond.com/docs/get-started/key-concepts/ · Key concepts, 'Business' section **Seat model** · No — Not a seat licence at all - teammates are free and unlimited, and the bill is driven by how many lead sources and phone numbers are connected, so there is no seat floor or ceiling. > There is no per-seat charge for teammates. Billing is based on the number of connected sources, not the number of people with access. > — https://auto-respond.com/docs/team-management/ · Section 'Billing and team size' **Pooled budget** · Yes — Everyone with access draws on the same business subscription and one shared usage meter; the documented $100 figure only triggers an invoice rather than capping what the team can spend. > the threshold is an invoicing trigger, not a spending cap, credit limit, or monthly maximum; > — https://auto-respond.com/billing-policy/ · Section 7, Usage billing and the $100 invoice threshold **Admin controls** · Yes — The agent is constrained by facts an admin approves - which services, prices, rules and answers it may use - plus per-source permission levels that decide who may edit those settings at all. > Control which services, prices, rules, and answers the AI may use. > — https://auto-respond.com/ · Auto Responder Features grid, 'Business Fact Guardrails' card **Audit log** · Partly — An Activity Log records source on/off changes and subscription events per business, and disabling AI identification in the voice greeting is recorded, but the vendor states it is not a full settings audit trail. > The log captures changes to active state and subscription events - it is not a full audit trail of every setting change. > — https://auto-respond.com/docs/activity-log/ · Section 'Scope', second bullet **Single sign-on** · No — The only sign-in the vendor documents is an individual email address and password per teammate; SAML and OIDC appear nowhere in the documentation, pricing or terms. > Each teammate signs in with their own email address and password. > — https://auto-respond.com/docs/team-management/ · Section 'Security', first bullet ## What happens to your code **Opt out of training** · Yes — Training on your call and lead data is off by default and contractual rather than a setting - it requires your written opt-in, and the same commitment is repeated in the privacy policy. > NZ Leads will not use Customer Personal Data to train an NZ Leads generalized AI model for unrelated customers unless Customer expressly opts in in writing. > — https://auto-respond.com/dpa/ · Section 4, Processing restrictions **Data residency** · Partly — Processing is in the United States and wherever subprocessors operate, with EU SCCs and the UK Addendum incorporated for transfers; no region selection or EU-only hosting is offered. > Customer authorizes NZ Leads and its Subprocessors to Process Customer Personal Data in the United States and other countries where they operate > — https://auto-respond.com/dpa/ · Section 15.1, International transfers, General **Getting out** · Partly — Leads export to CSV and recordings can be downloaded call by call, and press-release credits do not expire, but there is no documented bulk export of transcripts and cancellation does not delete data immediately. > Export CSV - downloads every lead matching your current filters as a CSV. > — https://auto-respond.com/docs/analytics-reporting/ · Top controls, 'Export CSV' bullet **Certifications** · Partly — GDPR, UK and Swiss transfers and CCPA service-provider duties are covered contractually by a DPA with the EU SCCs, but the vendor explicitly claims no certification or completed independent audit. > Nothing in this DPA represents that NZ Leads holds a particular certification or has completed a particular independent audit unless NZ Leads confirms that status separately in writing. > — https://auto-respond.com/dpa/ · Section 7, Security, closing paragraph ## Provenance Every fact above comes from the vendor. Nothing here is independently corroborated yet. ## Coverage elsewhere None recorded. Every fact above comes from the vendor and is not independently corroborated yet. ## What changed - 2026-09-11 — Capabilities · Commercial terms · Free tier - 2026-09-02 — Capabilities · Commercial terms · Free tier - 2026-08-31 — Capabilities · Commercial terms · Free tier · Price - 2026-08-23 — One-liner · Summary - 2026-07-30 — One-liner · Summary ## Correcting this record Anything marked "Not established" is a question we have not answered — it is not a No. If you represent Auto-Respond, or anything here is wrong or out of date, you can supply it with a source: the operations this site accepts from an agent are listed at https://theagentsindex.com/.well-known/agents.json — that catalogue needs no key to read, and each entry states its own auth. Supplying evidence is free. Nothing that can be bought here changes what a field says, whether a gap is closed, or how this record is scored.