Burst Desk - what a rate limit actually admits, as opposed to what it says

Paste your rate limits as they are actually configured, plus the traffic they face and what the service behind them can take, and work the whole thing in one sitting: what each limit ADMITS rather than what it says. Five lanes over one limit sheet - turn an API and its traffic into a sheet, read every gap with the mechanism that opens it, put every rule beside an exact sliding-window-log reference and get the boundary test that finds each one, work out what the per-key multiplication puts in front of the backend, and triage what a different algorithm closes, what a different number closes, what only the caller closes and what nothing closes because the property is the point. A free in-browser engine does the arithmetic before anything is spent - a fixed window admits exactly 2N at the boundary because two adjacent counters each admit N; a token bucket admits B + floor(r x elapsed), so its capacity arrives in one instant; 100/min across 8,000 keys is 13,333/s against a backend that takes 800/s; a concurrency limit is c/S and not a rate; a debounce is a step at the delay, so a thousand events 499ms apart under a 500ms delay fire once. It reads a sheet, not a gateway, and generates no load. Derived from the inngest-flow-control skill in github.com/inngest/inngest-skills, the rate-limiting skill in github.com/harperaa/secure-claude-skills, the marketplace-rate-limiting skill in github.com/vtex/skills and the tanstack-pacer skill in github.com/tanstack-skills/tanstack-skills. Not affiliated with or endorsed by their authors.

Every public app is built from a security-scanned skill and must pass a clean scan — skill and frontend — before it can be listed. Have a skill of your own? Turn it into an app — or read the step-by-step walkthrough.