Launching today
BillCurfew
Per-key AI spend alerts and a kill switch, no proxy
1 follower
Per-key AI spend alerts and a kill switch, no proxy
1 follower
Watches OpenAI, Anthropic & OpenRouter spend per API key. Polls about every 5 minutes, emails when a key passes a daily cap or spikes, and gives every plan a manual kill switch. Not a proxy, no code change, prompts never touched. Free for one key.
Free Options
Launch Team
Render Sandboxes: a computer for every agentGive every agent its own machine. Starts in under a second
Promoted
Hey Product Hunt 👋 I'm Jarrett, a solo founder in Orem, Utah.
BillCurfew gives your AI API keys a bedtime. Connect an OpenAI org admin key, an Anthropic Admin API key or an OpenRouter management key, then set a daily dollar cap and a spike rule for each key. It checks usage about every five minutes and emails you when a key goes over. If something looks wrong, one click kills the key. That works on every plan, including Free.
Why now: in October at least nine AI developer platforms disclosed incidents (Wiz calls it "Black October": https://www.wiz.io/blog/black-oc...), and Northflank users reported stolen AI keys being used. A leaked key or a looping agent shouldn't be something you first learn about from the invoice. Our breach checklist: https://billcurfew.com/guides/northflank-breach-ai-api-keys
What it isn't: it's not a proxy, so it reacts on the next poll instead of blocking the next request. It works with OpenAI, Anthropic and OpenRouter only; other providers like xAI, Gemini and DeepSeek aren't supported.
Free covers one key and doesn't need a card. Solo ($19), Team ($49) and Agency ($149) each come with a 14-day trial. We're at zero customers, so honest feedback helps a lot. What would make you trust a tool like this with an admin key?