WABridges
AI agents

Support customers where
they already are.

Your customers live on WhatsApp: let them message you there. Your backend receives every conversation as webhooks and replies via API. Own the entire flow, no third-party inbox, no per-agent fees.

Most support tools charge per seat, per conversation, or both. WABridges is different: you own the WhatsApp layer, and you build the handling logic in your own backend, or plug it into any helpdesk you already use.

Receive every customer message as a webhook POST. Reply with a single API call. Run it on your infrastructure, with your data, at your scale.

Four steps. One loop.

We keep the number online and hand you every event. You write the part in the middle.

1
Provision a support bridge

Create a bridge for your support channel. You'll get a customer_ref (e.g. "support") you use in every API call.

2
Set a webhook URL

Configure the URL where WABridges should POST inbound messages. Every message your customers send hits that endpoint as JSON.

3
Process and reply

Parse the webhook payload, look up the customer, route to an agent or bot, and reply via the send API. The whole loop is under your control.

Scale across products

Provision multiple bridges: one per product, one per region, one per support tier. All route to the same webhook or different ones.

The same three calls. Different jobs.

Every loop below is the one above with your own logic in the middle.

💬 Inbound support
  • Customer sends a question
  • Webhook hits your backend
  • Bot or agent replies
  • Full thread history in your DB
🤖 AI-assisted support
  • Pipe messages to GPT/Claude
  • Auto-triage by intent
  • Draft replies for agents
  • Auto-resolve common questions
🔔 Proactive outreach
  • Follow-up after ticket closed
  • NPS survey via WhatsApp
  • Renewal reminders
  • Onboarding check-ins
📊 Multi-number routing
  • One bridge per product line
  • One bridge per region
  • Route by customer_ref
  • Shared handler or separate

Paste into Claude, ChatGPT, or any AI assistant

Copy this prompt, paste it into any AI assistant, get working code in your language.

prompt
I'm building a WhatsApp customer support channel using WABridges.

My stack: [Express/Node.js: update this to your language/framework]
My setup:
- WA_API_KEY env var has my WABridges API key
- WA_WEBHOOK_SECRET env var has my webhook signing secret
- Bridge customer_ref is "support"

I need:
1. A POST /webhook endpoint that receives inbound customer messages
2. Verify the webhook signature using the X-Wa-Signature header
3. Send an immediate acknowledgement back to the customer by name if possible
4. Store the message for later handling (console.log is fine for now)

Please read the WABridges API docs and webhook reference before writing:
https://wabridges.com/llms-full.txt

Write production-quality code with error handling and signature verification.

One number, one price, no approval queue.

What you get on the WhatsApp side, whatever you build on yours.

Every inbound message delivered as a webhook, no polling, no delays
Reply via REST API from any backend, helpdesk, or AI model
No per-seat or per-conversation pricing: flat $5/bridge/month
Webhook signing: verify every inbound message is genuinely from us
Run multiple bridges for different products or support tiers
Own your conversation data: nothing goes through a third-party inbox