Still don’t see your answer? Reach out — our team gets back to you in under an hour on business days.
Fetchply is an AI chatbot that sits on your website and answers customer questions around the clock. It learns from your help center, product catalog, and policies, so it gives accurate, cited answers instead of generic ones.
Drop in a URL, train on the content you already have, and ship an AI agent that bridges the gap between a doc and a checkout. Two minutes from signup to a live widget.
Register any HTTPS endpoint once, and your chat agent calls it mid-conversation to fetch real-time answers. Order status from your own backend, live stock counts, account details: the agent pulls the answer, then replies in plain language.
custom_order_status200 OKOrder #6102 shipped on May 20 via USPS and is out for delivery today.
custom_inventory_check200 OKYes, fourteen in stock in size M right now. Want me to hold one for you?
Register any HTTPS endpoint once. The agent calls it mid-conversation for live answers: order status from your own backend, real-time stock counts, account details. Not stale crawl data.
Calendly booking links, Cal.com available slots, and calendar scheduling — ready to go with one credential paste. No URLs to copy, no parameters to define, no response mapping to figure out. The preset handles every detail and your agent starts calling it on the next turn.
List open times for an event type
Share your scheduling link on request
Write a one-line description and a call-when hint. The model reads both during the chat, then decides per turn: skip the call if the knowledge base covers it, fire the endpoint if the question needs live data.
Every call goes through an SSRF guard: HTTPS on port 443 only, private and metadata IPs rejected, no redirect following. Bearer tokens and API keys are encrypted at rest and never shown again.
Hand the full JSON to the model and let it write the reply. Or write a one-sentence template once, fill fields by path, and let the model stay out. Change the mode anytime.
Order {order_number} is {status}, shipped via {carrier}.
Pick GET or POST, paste your HTTPS URL, and name the parameters the agent fills before calling. Add a bearer token or API key header if the endpoint needs it.
Write a one-line description and a call-when hint. The model reads both and decides per turn whether the question needs live data from this endpoint.
Run a test call with sample parameters, preview what the agent sees, then publish. The endpoint goes live on your next turn.
Short, plain answers for store owners and developers weighing whether Functions fits their setup.
Register an endpoint, write a one-line description, and your agent starts pulling live answers from your own backend on its very next turn.