The Larry Effect: When Distribution Becomes a System You Build
Marshall from MDN Labs shared a clip about Oliver Henry's OpenClaw agent "Larry." The claim: Larry drove 8 million views in one week by running content distribution on autopilot. The skill is now free.
People are treating this like a novelty. It’s not. It’s a preview of where content goes when distribution becomes an operator problem.
Distribution Is the Real Bottleneck
Most people think content is the bottleneck. It isn’t.
The bottleneck is consistency: shipping every day, across multiple channels, with feedback loops that compound. Humans are bad at that. Not because they’re lazy — because they’re variable. Mood, energy, travel, deadlines, life.
An agent doesn’t have moods.
So the obvious move is to put an agent on the repeatable parts of distribution:
- formatting and repurposing
- scheduling
- posting
- responding and routing leads
- tracking what worked
- producing the next iteration from the data
Larry is one instantiation of that idea.
The "Free Skill" Is Not the Point
The skill being free is a nice gesture. The important thing is what it enables: a solo operator can now build a distribution machine that looks like a team.
This is the same pattern we’ve seen in software:
- in 2005, you needed a team to deploy
- in 2015, you needed one person + cloud
- in 2026, you need one person + agents
The cost of execution collapses. The value shifts up the stack.
Where the Moat Actually Is
If everyone can download the same skill, the moat isn’t the skill.
The moat is:
- the workflow you built around it
- the memory system that keeps it aligned
- the feedback loop that turns performance into improved output
- the distribution surface area you own (email list, audience, channels)
An agent can post. It can’t decide what you stand for.
Operators win when they combine:
- a clear point of view
- a repeatable system
- an agent layer that runs the system consistently
What I’m Building
This is why I’ve been treating distribution like infrastructure. If you’re building products in public, you don’t need a "content calendar." You need a machine.
BlackOps is my attempt to make that machine boring and reliable: publishing, promotion, feedback loops, and an agent layer that keeps it moving even when I’m offline.
Larry is proof that this isn’t theory. It’s already happening.
The next step is obvious: every serious operator will have a distribution pipeline. The only question is whether you build it before everyone else does.
Bookmark reference: https://x.com/mdnlabs/status/2023781809687814550
I wrote this post inside BlackOps, my content operating system for thinking, drafting, and refining ideas — with AI assistance.
If you want the behind-the-scenes updates and weekly insights, subscribe to the newsletter.
