◉CLAUDE LABJP
●WATERMARK — Since 11/16, text from Opus 4.6, Sonnet 4.5 and other models carries a watermark; non-default temperature and similar settings may drop it●CC 2.1.325 — Claude Code is still on 2.1.325 (11/13), with no newer release yet●SONNET 4.5 — Retires on the Claude API on 11/30, 13 days left. Sweep model IDs in code and saved settings in one pass●AUTOMATION — A user reports recurring AI jobs set up without run counts or a spending cap, burning through a weekly allowance●CONNECTOR — If connectors vanish only in new sessions, compare the /mcp list against an older session first●NEW — Cowork Feels Worse? Run the Same Request in Three Fresh Sessions●WATERMARK — Since 11/16, text from Opus 4.6, Sonnet 4.5 and other models carries a watermark; non-default temperature and similar settings may drop it●CC 2.1.325 — Claude Code is still on 2.1.325 (11/13), with no newer release yet●SONNET 4.5 — Retires on the Claude API on 11/30, 13 days left. Sweep model IDs in code and saved settings in one pass●AUTOMATION — A user reports recurring AI jobs set up without run counts or a spending cap, burning through a weekly allowance●CONNECTOR — If connectors vanish only in new sessions, compare the /mcp list against an older session first●NEW — Cowork Feels Worse? Run the Same Request in Three Fresh Sessions
Articles/Claude Code
⟐ Claude Code/2026-06-14Advanced

Pacing Non-Rollover Monthly Credits: A Burn-Rate Scheduler That Avoids Both Early Exhaustion and Wasted Balance

Non-rollover monthly credits punish you for spending too fast and for spending too slow. Here is the design of a scheduler that derives a daily burn rate from remaining balance and days left, throttles headless runs automatically, and the real numbers from running it on a personal automation setup.

Claude Code286Billing4Automation44SchedulerCost Optimization9

✦ Premium Article

I stopped mid-sip when the start-of-month summary printed. The previous billing cycle, I had left 18% of my reserved credit completely unused as the month rolled over.

If it had carried over, I would not have cared much. But headless runs after the June 15 change do not roll over. Whatever you do not spend vanishes the instant the month flips. Spend too eagerly in the first half, and the back half lands you on full API-rate metered billing.

Running several sites on automation as a personal developer, this asymmetry — losing money whether you overspend or underspend — wears on you quietly. This article shares the design of a burn-rate pacing scheduler that recomputes the "speed you're allowed to spend at" every single day, from remaining balance and the calendar.

The morning I found 18% sitting unused

What I run is a set of scheduled headless tasks — non-interactive claude -p style executions, staggered across the day, several per day.

The trouble was that token consumption per task is anything but uniform. A short integrity check finishes in a few thousand tokens; a long-form generation task can reach hundreds of thousands in a single run.

Early in the month I'd feel I had room and run generously; as month-end neared I'd unconsciously hit the brakes, afraid of running short. Running on human intuition is exactly what produced that 18% of unused balance. I had erred so far toward safety that I never put the reserved budget to work.

With non-rollover monthly credits, that gut-feel operation becomes a direct loss. A balance that is designed to disappear has to be spent deliberately — but without starving out. That calls for numbers derived from balance and date, not a feeling.

The two-sided loss that non-rollover credit creates

First, the constraints. After June 15, the Agent SDK, headless claude -p, GitHub Actions, and third-party agents moved to a separate monthly credit pool, distinct from subscription usage limits. The pool is $20 (Pro) / $100 (Max 5x) / $200 (Max 20x) depending on plan, overage bills at the full API rate, and — critically — it does not roll over.

That non-rollover rule produces a loss in two directions.

One is early exhaustion. If your burn rate runs too hot in the first half, every run after the pool empties flows to metered billing. A cost you thought you had fixed creeps toward open-ended by month-end.

The other is leftover waste. Lean too far toward safety and the balance evaporates unused at month-end. That's the one I hit. 18% of a $200 pool — about $36 — thrown away every single month.

The ideal is a steady pace that lands the balance near zero exactly at month-end. But real tasks vary in consumption, so a fixed schedule can never match it. That's precisely why you need a mechanism that measures the pace and nudges it daily.

✦

Thank you for reading this far.

Continue Reading

What follows includes implementation code, benchmarks, and practical content we hope you'll find useful. This site runs without ads — server and development costs are supported entirely by members like you. If it's been helpful, we'd be truly grateful for your support.

WHAT YOU'LL LEARN
✦A pacing formula that derives today's spend ceiling from remaining credit and days left, with the complete TypeScript implementation
✦Pre-estimating the token cost of each headless run so the queue stops before it overshoots — which cut my end-of-month waste from 18% to 2%
✦Decision rules for using up a non-rollover, full-API-rate credit pool without starving high-priority tasks
Secure payment via Stripe · Cancel anytime
✦

Unlock This Article

Get full access to the rest of this article. Buy once, read anytime. This site is ad-free — your support goes directly toward keeping it running.

or
Unlock all articles with Membership →
Share

Thank You for Reading

Claude Lab is ad-free, supported entirely by members like you. We publish practical guides daily with implementation code, benchmarks, and production-ready patterns. If you've found it useful, we'd love to have you on board.

  • ✦Copy-paste ready implementation code
  • ✦New advanced guides published daily
  • ✦$5/mo or $10 for lifetime access
View Membership →

Related Articles

⟐ Claude Code2026-09-01
Drop the quotes on a heredoc and the prices in your log quietly change
An unquoted heredoc runs the variables and backticks inside its body. Here are the four ways my log got rewritten, how the three quoting forms compare, a safe placeholder-and-sed pattern, and a small script for auditing what you already have.
⟐ Claude Code2026-08-22
Handing a long job to another session — and the completion marker for when the notification never arrives
How to use notify_when_idle to hear when another Claude Code session finishes, and a small completion marker that keeps you from waiting forever when the notification is dropped.
⟐ Claude Code2026-08-03
Existence Checks Pass, Writes Fail — Probing Capabilities Before an Unattended Run
A directory existing and a directory being writable are two different facts. Measured results from five broken-environment cases, why static clues cannot predict the worst of them, and a capability-probe preflight for unattended Claude Code runs.