Text store commands to an agent
Use one agent to run store actions from Telegram while another answers customers on WhatsApp.
- What they did
- Claw Boutique is an AWS sample built around two agents that share a single Store API. A buyer-facing Strands agent on Bedrock AgentCore, using Nova Lite, handles WhatsApp questions and order updates. An OpenClaw agent on EKS watches Telegram for the owner’s instructions, flags low stock and poor reviews, and executes store actions such as restocking, replying, or starting a refund flow.
- How it went
- The walkthrough shows an order moving across the web shop, WhatsApp, email, Telegram, and an admin dashboard. It does not report any speed or revenue figures, and it is presented as a reference setup rather than a live production system.
- Tips
- A single CDK deploy sets up the stack, but the first deployment can take about 25-30 minutes. You also need a WhatsApp Business account, a Telegram bot token, and a verified SES sender address, and the EKS cluster plus RDS database keep running until you tear everything down.
Try it yourself with OpenClaw
Act as my store operations agent. I will message you through [Telegram], and you should translate my text into the right action against my [store API]. Handle requests like [example command 1], [example command 2], and [example command 3]. Before you do anything that changes data, sends a message to a customer, issues a refund, or deletes information, first tell me exactly what you plan to do and wait for my approval. After each action, report back with the result and any follow-up I should review.
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.
Related agent uses
Run a small agent fleet from Notion
They used Notion as the control panel for 18 OpenClaw agents, with tasks, logs, alerts and configs all flowing through one workspace.
Turn off thinking in cron jobs that need to send digests
A developer fixed an OpenClaw cron setup so its article digest actually reached Telegram.
Set up a Slack-controlled agent with memory and a dashboard
They ran OpenClaw on a separate MacBook Air, used Slack for most instructions, and tracked its work in daily notes and a live dashboard.
Use an agent to manage roadmap, support, calendar, and tasks
A developer set up OpenClaw to sort support mail, update a roadmap, handle calendar items, and keep track of todos and sponsorships.
Automate inbox checks, research digests, and site audits
Jenny Ouyang uses two OpenClaw agents with scheduled jobs to scan email, draft research digests, and monitor a website.
Create a local control panel for day-to-day operations
A solo business owner used OpenClaw to assemble a local dashboard for approvals, scheduled work, and workflow tracking, and it replaced several subscriptions.