AI Customer Support Agent for 24/7 Automated Support

reply_with_response replies using specific fields from the API's response; reply_with_static_reply sends a fixed message regardless of what the API returns; no_reply calls the API silently, with nothing sent to the customer. This is one of the most flexible ai agent integrations available in the system prompt.

##save_custom_field## : 42 writes a value to a custom field on the contact — useful for capturing plan tier, preferred language, or order size directly from the conversation, without a human typing it in manually.

Yes — ##skip_reply## tells the agent not to respond to a vague or irrelevant message, and ##call_http_api## : 5 | no_reply lets it call an external system in the background with no customer-facing message at all.

Instruct the agent to use ##block_subscriber## only when a message clearly matches spam patterns you define in the prompt, and ##skip_reply## for anything merely low-value — that distinction keeps a real AI agent for customer service from mistakenly blocking a genuine customer.

Yes — ##add_note## : Customer wants a callback saves an internal note visible only to your team in the shared inbox, and ##follow_up## : 24 sets a staff reminder a set number of hours from now.

Routing rules match the topic of an incoming message — billing, product questions, technical issues — to the right specialist. An agent you've created but not enabled under Agent Configuration simply won't be part of that routing.

The current agent can trigger ##transfer_agent## : 18 to hand the conversation to another AI agent — for example, a Support Agent transferring a pricing question straight to Sales — without the customer having to repeat their issue.

##transfer_agent## moves the conversation to another AI specialist. ##assign_human_member## or ##assign_human_role## hands it to an actual person in your shared inbox — one specific teammate, or a round-robin across a team role — typically once the customer explicitly asks for a human.

Yes — ##mark_as_solved## closes the conversation once the agent judges the issue resolved, keeping your inbox from filling up with tickets that were actually answered.

Both. It's built as ai chatbot software for customer support out of the box, but the same agent framework — system prompt plus actions — configures just as easily into a Sales Agent or a Lead Qualification Agent. For non-AI journeys, the visual flow builder handles scripted sequences alongside the agent.

Under Knowledge Sources (Media), not inside the prompt — pasting image URLs directly into the system prompt isn't supported. The agent pulls from that media library and sends the right image when a customer asks for photos.

No — bind the store once with ##shopify_store## or ##woocommerce_store##, then every order-lookup, product-recommendation, and cancellation action becomes available. Other store actions simply won't save until that binding is in place. See the setup for Shopify with WhatsApp and WooCommerce with WhatsApp.

Yes — labels, sequences, custom fields, and follow-up reminders all work alongside the agent, so the broader set of AI tools for customer support covers CRM organization and internal workflow, not just the conversation itself. WhatsApp broadcasting handles the proactive side.

Yes — many teams start with a single AI customer service agent answering FAQs, then add order lookups, HTTP API calls, and a second specialist agent once the first one is proven out.

It escalates — using ##assign_human_member## or ##assign_human_role## to hand the conversation to your team, with ##add_note## capturing context so the person picking it up isn't starting cold.

Yes, when it's configured with clear escalation rules. Treat customer support agent ai as the first line of response — resolving the repetitive, well-documented questions instantly — while genuinely complex or sensitive cases route straight to a human, which is exactly what the routing and handoff actions above are built for.