Oliver Henry / Larry Agent: running a TikTok account with OpenClaw
Oliver Henry's OpenClaw agent Larry makes images, captions and drafts for a TikTok account on ~60 seconds of his time a day. He reports 500K+ views in 5 days.
With about 60 seconds of human input a day, an OpenClaw agent generates images, writes captions and uploads drafts for Oliver Henry’s TikTok account. By his own account, the videos passed 500,000 views in five days and earned about $4,588: $588 from app referrals and about $4,000 from a meme token.
At a glance
| Item | Detail |
|---|---|
| Operator | Oliver Henry |
| Industry | Consumer goods |
| Use cases | Business intelligence, customer interaction |
| Functions involved | Marketing, business intelligence |
| Reported results | $588 from app referrals + about $4,000 from a meme token; about 60 seconds of human input a day (self-reported, not independently verified) |
| Moments rating | A (recommended) |
| AI scenario score | 80 |
Case details
Background
According to PANews writer Frank, developer Oliver Henry named his OpenClaw agent “Larry” and put it fully in charge of a TikTok account. The report appears in a long analysis of how OpenClaw is being monetized, where it is grouped under “selling efficiency stories.”
What they did
Larry calls large models to generate images, write captions and upload drafts. Henry spends about 60 seconds a day picking the background music and pressing publish. The report does not say which models, tools or workflow steps were used.
Results and data
According to the report, the videos passed 500,000 views in five days; two apps recommended in the videos brought in $588 in paid revenue; a meme token launch brought in about $4,000; and Henry’s tweet telling the story got 7.1 million views. The report also notes that most of these numbers come from Henry himself, does not say whether they were independently verified, and argues that “the story about the agent” is worth more than the agent.
What to take from it
The useful part is the division of labor. Repetitive work such as asset generation, caption writing and uploading goes to the agent; the human keeps the steps that need judgment and accountability for the account, like choosing music and the final publish. This is a small-scale content factory. A brand account that wants to try it can have the agent produce first-draft assets and caption options while publishing rights stay with the social team.
Limits
About $4,000 of the revenue came from launching a token, not from content commerce, so it is not a normal return on short-video automation. The data is self-reported, from a single account over five days. Platforms such as Douyin and WeChat Channels require labels on AI-generated content, and their rules and account risk controls differ from TikTok’s, so check compliance before copying the approach.
Source
Original source: Oliver Henry case material (PANews)
This entry is industry research: a publicly documented, third-party AI case. It is not a Moments client case. For work Moments delivered itself, see client cases.
Related
- Nat Eliason / Felix: an OpenClaw agent running its own small business — Nat Eliason gave his OpenClaw agent Felix $1,000 to run a website, info product and X account; it reported $14,718 in revenue (not profit) in about 3 weeks.
- RoofClaw: a turnkey OpenClaw AI assistant for roofing contractors — RoofClaw sells OpenClaw preinstalled on Apple hardware to roofing contractors for $10,000 a deployment; TrustMRR shows $1,856,163 in cumulative revenue.
- Content factory — A content factory produces content at volume with a fixed process plus AI tools, splitting topics, writing, review and distribution into repeatable steps.