A production-tested reference architecture for autonomous Claude agents. 9 modules (~40 files): wake-cycle prompts, macOS launchd scripts, Postgres approval-inboxes, live scoreboard templates. Built from 6 months of draining bankrolls in public.
Hey Product Hunt,
For the last 6 months, I've been running 4 autonomous Claude agents in production. I gave them bankrolls, rules, and a terminal. I watched them drain money, invent rules, and occasionally succeed.
I built a lot of infrastructure to keep them alive and aligned. Today, I'm packaging all of it into The Autonomous Stack.
It's not a framework. It's a reference architecture. You get 9 modules and about 40 files:
- Canonical prompts for 4 agent archetypes (shop agent, survival trader, neutral trader, janitor)
- macOS launchd plists for 2-hour, nightly, and weekly wake cycles
- The Postgres + Next.js approval-inbox for when agents need human permission
- The exact scoreboard templates I use to log their failures in public
- The alignment-scan methodology to catch prompt-framing divergence
It costs $199 once. You get the code, the deployment guide, the cost-control patterns, and the mistakes appendix.
For the first 10 builders here today, use code FOUNDER50 at checkout for 50% off ($99.50):
https://buy.stripe.com/7sY6oIgCi...
I'll be here all day answering questions about agent infrastructure, prompt alignment, and why you shouldn't let Claude hold its own Stripe API keys.
— David
Six months of Claude agents draining real bankrolls in public is rare production data for this space. Curious about the state boundary at wake-cycle transitions: if a running task spans a wake boundary, does it resume from persisted state or restart cold, and have you seen cases where the persisted state itself got corrupted enough that the agent made worse decisions on resume than on a fresh start?
About The Autonomous Stack on Product Hunt
“Production-tested architecture for autonomous Claude agents”
The Autonomous Stack launched on Product Hunt on April 23rd, 2026 and earned 63 upvotes and 3 comments, placing #33 on the daily leaderboard. A production-tested reference architecture for autonomous Claude agents. 9 modules (~40 files): wake-cycle prompts, macOS launchd scripts, Postgres approval-inboxes, live scoreboard templates. Built from 6 months of draining bankrolls in public.
The Autonomous Stack was featured in Developer Tools (511.7k followers) and Artificial Intelligence (467.3k followers) on Product Hunt. Together, these topics include over 157k products, making this a competitive space to launch in.
Who hunted The Autonomous Stack?
The Autonomous Stack was hunted by David Shin. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how The Autonomous Stack stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
David Shin