Run a side project with an always-on agent

They used Clawbot on a Mac mini to handle recurring content, research, analytics, and code tasks, with a human still approving the results.

Done with OpenClaw· Intermediate· Apr 2026

What they did
The poster kept Clawbot running on an always-on Mac mini and connected it to the systems they already use for work, records, metrics, and site operations. They assigned it recurring SEO work, social drafts, lead-finding, analytics spot checks, and programming tasks. The agent worked from both scheduled daily runs and Discord-based triggers; for software changes, it drafted updates and opened pull requests, then paused for the founder’s review.
How it went
This let the side project make progress every day instead of slowing down when the founder was busy. They reported more organic traffic after regular publishing, a few signups from posts the agent wrote, and a weekly savings of about 3–4 hours on lead research. Social responses were still scarce, and they kept reviewing everything before it was published or merged.
Tips
Getting it set up took a couple of weekends. Even at about $1k MRR, they still checked all SEO output by hand to reduce mistakes.

Try it yourself with OpenClaw

Act as the operator for my side project, [project name], which helps with [one-line description]. Every [day/time], draft [content type] and save it in [review location] for my approval. Every [day/time], inspect [analytics/report source] and give me a brief list of any notable changes. When [event] occurs, start [coding task] and prepare a pull request, but do not publish, send, delete, or merge anything until I say it’s okay.

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.