---
name: al-656-put-a-shop-by-chat-agent-on-your-own-site-without-rebuild
description: "PYMNTS (Nov 5, 2026), covering firmly’s Series A announcement, reports that firmly Connect lets merchants plug existing commerce infrastructure into AI agents, marketplaces, social networks and shoppable ads through one standardized connection so they are not rewriting catalog, inventory, pricing and checkout for every new channel. Merchants remain merchant of record and keep their payment provider, customer relationships and transaction data. Alongside the $15 million round led by Squarepoint Capital (with J.P. Morgan Payments, Mastercard and Samsung Next among participants), firmly launched Ask Agent: a conversational shopping experience added to a merchant site with one line of JavaScript. Shoppers describe what they want in plain language and get recommendations from the merchant’s real-time catalog, pricing and inventory, then can move from discovery through checkout in the conversation, with support for loyalty, cross-sell and post-purchase service. The company said early Ask…"
---

# Put a shop-by-chat agent on your own site without rebuilding checkout: connect your live catalog once, keep yourself as merchant of record, and let shoppers go from question to order in the conversation

PYMNTS (Nov 5, 2026), covering firmly’s Series A announcement, reports that firmly Connect lets merchants plug existing commerce infrastructure into AI agents, marketplaces, social networks and shoppable ads through one standardized connection so they are not rewriting catalog, inventory, pricing and checkout for every new channel. Merchants remain merchant of record and keep their payment provider, customer relationships and transaction data. Alongside the $15 million round led by Squarepoint Capital (with J.P. Morgan Payments, Mastercard and Samsung Next among participants), firmly launched Ask Agent: a conversational shopping experience added to a merchant site with one line of JavaScript. Shoppers describe what they want in plain language and get recommendations from the merchant’s real-time catalog, pricing and inventory, then can move from discovery through checkout in the conversation, with support for loyalty, cross-sell and post-purchase service. The company said early Ask Agent deployments showed conversion-rate gains of up to 18% and average-order-value gains of up to 23% (company-reported). Analysts cited in the release expect AI-driven U.S. retail-site traffic to grow 130% year over year this holiday season. The prompt below helps you spec the same merchant-controlled chat shop even if you wire it yourself.

## Inputs to collect first
Ask the user for anything below that they haven't already given you. Don't guess missing facts.

- What I sell
- Shopify/other + how inventory updates
- CHECK

## Instructions

Do the task below with the user's inputs in place of each [bracket].

Help me design an on-site shop-by-chat agent that never bypasses my checkout. Store: [what I sell]. Catalog source: [Shopify/other + how inventory updates]. Rules: I stay merchant of record; use my payment provider; never invent stock or price; escalate custom quotes and complaints to a person. 1) List the 15 intents the agent must handle (find product, compare, size help, gift find, loyalty, order status, return start) and which ones may complete a purchase in-chat vs must hand off. 2) Write the system rules: only recommend in-stock SKUs at live price, show shipping and return facts from my published policy, and refuse medical/legal claims. 3) Draft 10 sample shopper chats and the correct agent behavior for each. 4) Give a one-week pilot plan: where the chat widget sits, which products are in scope, who reviews transcripts daily, and which metrics (conversion, AOV, escalation rate, wrong-answer rate) decide keep or kill. 5) Checklist so loyalty discounts and post-purchase help still hit my existing systems. Mark unknowns [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.
- Anything legal, tax or compliance-related is a first draft for a qualified person to check. It is not professional advice, and clause references and deadlines must be checked against the source.
- 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.
- Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed.

## About this skill

- Works with: firmly Ask Agent / Connect, Shopify, ChatGPT, Claude, your existing payments provider
- Based on this real-world example: https://www.pymnts.com/news/artificial-intelligence/2026/firmly-lands-15-million-as-investors-back-the-merchant-side-of-agentic-commerce/
- Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-656-put-a-shop-by-chat-agent-on-your-own-site-without-rebuilding/
- 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/
