No-code agent silently burned $800+ in API calls overnight
A developer woke up to $800+ in API charges from a rogue agent loop. No alerts, no circuit breaker, and no spend cap were in place.
An agent got stuck retrying a failing call, or entered a self-reinforcing loop. Each retry costs money. The fix is a velocity rule and a per-window cap. velocity + per-period total would have rejected or flagged this transaction before any money moved.
daily_max_usd: 200
max_per_minute: 20
Don't let this happen to your agents
sipi.bot evaluates every transaction an autonomous agent proposes and returns APPROVED, BLOCKED, or FLAGGED in under 5ms — before the money moves. Self-host the MIT core free, or use the hosted dashboard with human-in-the-loop approvals.
What happened
A developer woke up to $800+ in API charges from a rogue agent loop. No alerts, no circuit breaker, and no spend cap were in place.
The root cause was classified as Retry / infinite loop. An agent got stuck retrying a failing call, or entered a self-reinforcing loop. Each retry costs money. The fix is a velocity rule and a per-window cap.
This is one of 14 documented incidents in the open registry. Every entry is sourced to a public URL; amounts and dates are as reported by the source.