Coding agents remove much of the friction from building and deploying software. Simon Willison argues that this also removes friction from accidentally creating systems that keep spending money while their owners sleep.

His proposed default is simple: usage-priced services should stop at a fixed dollar limit, not merely send an email after the threshold has been crossed.

  • Hard caps should be on by default; accepting unlimited overages should require an explicit opt-in.
  • A service failure is usually less damaging than an unexpected bill for thousands of dollars.
  • AWS has started rolling out project spending limits, though only through a limited new account experience.
  • Google Cloud now offers monthly caps for specific services.
  • Agents should favor providers with enforceable caps and warn people before recommending uncapped infrastructure.

The important distinction is between monitoring and control. A warning reports that spending has gone wrong; a hard cap limits the damage. That matters more as agents can create hosted applications, call paid APIs, and add storage or compute without someone supervising each step.

This is not only an AI-token problem. Agents expand the number of metered systems a small project can touch, so the safe boundary belongs at each provider—not in a prompt asking the agent to behave.

The 71-comment thread on Hacker News adds a useful correction, an operating example, and the strongest objection.

What the thread adds

  • modeless — Google Cloud’s feature is much narrower than the headline suggests: “Literally only works for four random services, unsupported for all the rest.” there_is_try adds: “The hard caps work on projects created in AI studio.”
  • zanecodes — client-side circuit breakers are not always practical because cloud pricing can be “incredibly complex or opaque,” while usage figures may not update fast enough to stop an expensive operation.
  • tkgally — a concrete agent-level fallback: “I tell Claude to keep to hard daily limits on OpenRouter spending for two long-running projects.” Once the limit is reached, the agent switches to work that does not require paid model calls.
  • twoodfin — the operational counterargument: “Never let me spend more than $X” can also mean “Shut down my business-critical app/service/solution at 2 am on a Sunday morning because Joel in IT forgot to plan for the new report runs.”

HN handles are pseudonymous and the site publishes no per-comment scores. The ordering is HN’s own ranking, so this is a slice of the thread, not a consensus.