Budget Tokens For Feedback Loops
budget-tokens-for-feedback-loops
Summary
Long-running agent work can exhaust context before validation, review, or handoff happens. Reserve space for feedback loops by summarizing durable state and avoiding unnecessary transcript growth.
Problem
Token spend is easy to treat as waste when it is measured like a software bill instead of as part of the cost of doing work. Under-spending on review agents, cleanup loops, evals, and verification can leave human attention as the bottleneck.
Preferred Move
Budget agent tokens for feedback loops that improve outcomes: validation, review, security scans, garbage collection, documentation checks, and harness evals. Measure the value of the loop by quality, throughput, and reduced human attention, not by raw token volume.
Tradeoff
Do not Goodhart token consumption. A token floor is useful only when it funds a loop with a clear purpose and evidence of value. If the loop stops improving outcomes, change or remove it.
Agent Instruction
When proposing an agent loop, state what signal the token spend buys and how its effectiveness will
be judged. Avoid treating token volume itself as success.Examples
Bad: token spend becomes the metric.
Run more agents until the weekly token budget is used.
Good: token spend funds a measurable feedback loop.
Run the nightly docs-drift and security-review agents. Track actionable findings, false positives,
and human review time saved.
References
| Source | Use | Note |
|---|---|---|
| OpenAI scarce attention | adapts |
Human time and attention are treated as the scarce resource. |