Set a work-finished WhatsApp ping with reply filtering

A user arranged for OpenClaw to alert his wife on WhatsApp after work ended, then separated her outgoing replies from the agents that should answer.

Done with OpenClaw· Intermediate· Oct 2026

What they did
He configured OpenClaw to send his wife a WhatsApp note when he wrapped up work. To make outbound messaging possible, her contact had to be approved in a way that also forwarded her responses into the system. He got around that by clearing the shared default model, attaching models only to the agents meant to respond, and adding a no-model "silence" agent as the destination for her number.
How it went
That setup prevented her messages from being handled as instructions. He said the method was a workaround and asked for a built-in per-contact option that would permit sending while blocking inbound processing.
Tips
If your agent platform has one default model for everything, check whether contact approval affects both directions. Until there is a native per-peer control, keep response-capable agents separate from sink addresses and test carefully before turning it on.

Try it yourself with OpenClaw

In OpenClaw, help me build a rule that sends [name] a WhatsApp update at [phone number] when I finish work, using [message text]. Before anything is sent, show me the configuration and wait for my approval. Also ensure that messages from [name] do not get handled by any active agent; if the platform cannot do that directly, outline a safe workaround and ask me to confirm it before applying it.

Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.

Get 3 agent uses like this every morning

With the day's top AI news. Free, and one click to unsubscribe.