Subagent review loops burn tokens on simple tasks
Subagent-driven development burns tokens and wall-clock time on simple tasks through repeated review loops.
Tier II · Emerging Problems that are still building up.
4 public reports · first reported 10 April 2026 · last reported 22 July 2026
What's going wrong
Users running subagent-driven development report that a full two-stage review on trivial, fully specified tasks costs many times the tokens of doing the work directly.
AI-assisted: written with the help of AI from the reports. The idea report has the full summary.
In the idea report
- The full summary.
- The products the 4 reports are about, and how many came each month.
- Related problems.
No business has been found with this problem yet, so the report has no contacts. All sales are final — no refunds. Pricing.