Four-agent LangChain loop runs 11 days and burns $47,000 while dashboards look healthy
Teja Kusireddy team (multi-agent research system) · Mar 23, 2026 · General agent
What happened
A multi-agent research system managed by Teja Kusireddy's team generated a $47,000 invoice due to a recursive communication loop. Two agents exchanged thousands of messages over eleven days without producing useful output, while dashboards showed healthy activity and normal latency -- the billing damage was surfaced only after the invoice arrived. The post-mortem identified two root causes: no per-agent budget ceiling, and no enforcement mechanism that would have terminated the session before the next API call.
Causal vector
Unbounded multi-agent communication loop with no token or cost ceiling and no pre-execution enforcement -- only asynchronous monitoring dashboards
Source
Reported by Waxell (citing primary post-mortem by Teja Kusireddy). Verified against the primary report.
This is the textbook runaway loop in a multi-agent setting. A velocity cap on calls per minute per agent and a daily total cap return BLOCKED on the call that would push spend over the ceiling -- stopping the loop cold at the infrastructure layer, not as a dashboard alert someone may or may not act on.
The six rule types that contain this class of failure
Per-transaction cap
Any single spend above your ceiling is BLOCKED before it moves.
Daily total
Cumulative spend across all agent calls, bounded per day.
Velocity limit
Stops runaway retry loops — the #1 cause of overnight losses.
Merchant allowlist
Only approved destinations can ever receive funds.
Category rules
Flag high-risk classes (crypto, infra, refunds) for review.
Approval threshold
Above a value, the action waits for a human.
Related incidents
OpenAI Codex stuck in a 9-hour compaction loop drained an entire weekly usage allowance
Codex user (OpenAI Developer Community) · Aug 25, 2026
Replit Agent's effort-based billing hits $50/hour as looping inflates billable actions
Replit Agent users · Jul 16, 2026
Stanford Digital Economy Lab: agentic coding tasks consume ~1,000x more tokens than chat, with up to 30x variance on the same task
Stanford Digital Economy Lab · May 5, 2026
Don't be the next entry
Every incident in this database is the result of trusting a prompt, a provider cap, or a human review cycle. sipi.bot replaces all three with one deterministic call. 85 documented failures, one control.