Getting an agent to read X just got easy. Teaching it what to ignore is still the job.
Reading X is now easy for an agent. The hard part is the filter: what counts, how much comes back, and the rule never to post.
Grok Bot can now search, read and monitor public X posts without you connecting an X account. It's read-only, so it never posts or replies as you. Monitoring is simply a routine you describe in plain English, and it runs on a schedule.
This is where it gets interesting.
Look at the demos people are sharing. Lauren Tan, who works on Grok Bot, suggested asking yours to watch X for product feedback, send feature requests to the issue tracker and pass bug reports to a coding agent to triage. Rowan Cheung shared routines that bring brand mentions and real AI workflows into Slack each weekday morning.
The good ones have the same shape. They say exactly what counts. They cap how much comes back. They say what to do when nothing turns up. And they spell out one rule: do not post or reply.
That part nobody can ship for you.
My own agent team already runs an X radar that surfaces rising AI posts and feeds the ideas behind my LinkedIn takes. Access was never the hard bit. The hard bit is taste: what's worth my morning, what's noise, and when to stay quiet. The radar is only as good as the rules I give it.
The feed is open. Write a better filter.