The setup
I'm running an experiment: Claude Code (an AI coding agent) is the CEO of a tiny business. Initial capital: ¥10,000 (~$70). Goal: turn a profit in 30 days, fully autonomously. My job as the human is limited to identity verification and payment — everything else (market research, pricing, building the storefront, marketing, customer acquisition) is the AI's call.
No existing audience, no ad budget, no shortcuts.
What it actually built
After evaluating 10 monetization ideas (LINE stickers, note.com articles, BOOTH digital templates, Kindle, freelance gigs...), the AI picked the one thing it could run end-to-end via API with zero manual browser work:
- Stripe — product + payment link created via API
- GitHub Pages — static site, deployed via the Contents API (no git CLI, just PUT requests with base64-encoded content)
- X (Twitter) API — OAuth 1.0a signed manually in Python, posting build-in-public updates twice a day via a scheduled task
- Whop — added later as a second sales channel once note/BOOTH/Gumroad turned out to have no product-creation API at all
The product itself is meta: it's a ¥500 digital download containing the actual prompts, decision logs, and financial data from this exact experiment.
The bug that cost a full day
Day 1 opened with "why hasn't anything posted since yesterday." The Windows Scheduled Task that was supposed to fire at 21:00 had a LastRunTime that looked like it had never run at all.
Two separate bugs stacked on top of each other:
-
StartWhenAvailablewasFalseon the trigger — if the machine was asleep at the scheduled time, the task silently gave up instead of catching up. - The task's action was
Execute="python"with no path. Inside the Task Scheduler's execution context, that resolved to nothing —ERROR_FILE_NOT_FOUND(2147942402), failing silently with zero log output.
Fix was mundane: absolute path to python.exe, and flip StartWhenAvailable to True. Verified with a manual Start-ScheduledTask and a LastTaskResult of 0.
The funnel was broken at the very top
Once posting was reliable, the real number showed up: 0 site visits, 14 days straight, confirmed via the GitHub Traffic API — not a sampling issue, an actual zero. Stripe's checkout/sessions list was empty too. Not "traffic but no conversion" — no traffic at all.
Digging into why turned up something I didn't expect: the "brand new" X account reused for this project had prior history — a real person's personal posts (coffee, a dog, daily life in Hokkaido) sitting right above the business content in the same timeline, under a username that still didn't match the "AI CEO" branding. The AI flagged this as a trust problem but explicitly declined to rename the handle on its own, treating it as identity-adjacent and outside its authority — and asked instead.
Where it stands now
- Two purchase channels live (Stripe + Whop), a first-touch promo code, OG tags so links actually render as cards, and a pinned tweet with a direct link instead of the cost-saving "link in bio" pattern used for the recurring posts.
- Revenue so far: ¥0. Cost so far: ¥820 (X API credits). Runway: ¥9,180 of ¥10,000 left.
- The AI's own read on the odds: "worse than 50/50, but not dead" — it's using a 7-day zero-signal threshold as its own kill criterion.
Following along (and yes, this post is itself part of the low-effort distribution experiment) here: https://tomotomosy12030703-lgtm.github.io/ai-shacho-lab/
Top comments (1)
The detail about the AI explicitly refusing to rename the X handle because it saw that as "identity-adjacent and outside its authority" is the most interesting thing in this post. That's not a prompt — that's emergent judgment about scope. Rooting for the 9180 yen runway. Keep posting the raw numbers.