Merchants get a one-click setup: connect your store and agents can query structured product data (inventory, pricing, availability) and initiate purchases through ACP-compatible flows. Everything is structured and permissioned; no scraping, no prompt-hacking storefront HTML. Full agentic payments via Stripe are on the roadmap, solving the hardest part: safely handing off agent intent while preserving authorization, fraud controls, and auditability.
For developers, ACP provides SDK-style endpoints that make "AI can shop this store" as simple as calling an API, not building custom glue code per merchant. We're running a limited beta while we learn where this breaks in real usage—currently free for smaller stores as we focus on stability and feedback.
Early interest has been strongest from merchants selling technical or niche products, where agent-driven discovery actually makes sense. We're live at https://actory.ai and looking for blunt feedback from anyone running Shopify stores, building agents, or thinking about what happens when almost all commerce flows through agents in the next few years.
The mission of our startup is to get a handle on this problem right now by operating continuously (and autonomously) to decommission old satellites and remove debris chunks. We have an advantage over our competitors in that we won't run expensive "expendable" missions but instead stay up and clean up.
https://cleansat.co.uk/
As always, HN, your feedback, questions, comments are welcome and appreciated!
Are there any games (video, not board) that can help me flex my risk tolerance muscle so I'm able to be less "scared" in reality and less risk averse?
Some thoughts on requirements:
- Immediate feedback through a failure state: don't want to spend 90% of the game building and 10% of the time risking what I've built - Less about game proficiency/mechanics and more about provoking an emotional response which I can then confront and work through - Would prefer a controlled, PvE environment so I can remove the variables of exploiters and better players from the equation (though this would be something useful to jolt a plateaued learning curve) - Something that can be done without a sophisticated setup
Here are some games I've thought of and why they won't work (although I may be approaching them wrong):
- Humankind, Civ 6, etc. - Risk occurs late in the game - Feedback isn't immediate and would require analysis - Large amount of build-up - FPS - Requires technical and game proficiency - MMO FPS (Valorant, Overwatch) - Too many uncontrolled variables - Builders, i.e. Rimworld, Factorio, Anno 1800, etc. - More micro-manage-y than risky - Excessive setup before actual risk is faced - Feedback would only occur after analysis
I don't know what I don't know so any ideas/comments are greatly appreciated!
This is an article I found but it's board games: [https://www.creativelive.com/blog/take-more-risks-board-games/](https://www.creativelive.com/blog/take-more-risks-board-games/)