They kept running into usage caps. Claude built a status line showing usage, reset time and context, plus a hook that catches spikes, like a long chat suddenly resending everything.
Result Spikes caught before they burn the quota
“Between the two I haven't hit usage caps in weeks.”
Spotted: found in the wild. Details are as reported by the source above and have not been independently verified.
Used Claude to write a plug-in that changes the bottom of the CLI to show the usage percentage, time until reset, context used, model and mode, with an indicator that changes colour when usage gets high. Paired it with a hook that watches for usage spikes, such as a long conversation suddenly resending lots of context every turn.
Between the two, they haven’t hit a usage cap in weeks.
Write a status line script for Claude Code that shows my current usage percentage, the time until my limit resets, how much of the context window is used and the model, and changes colour above [threshold] percent. Then write a hook that warns me when one turn uses far more than my recent average, so I can start a fresh session before burning my allowance.
A starting prompt written by Agents in the Wild. Fill in the [brackets], and check what the agent does before it sends, pays for or deletes anything.