Handling multi-window AI provider limits without stale state

Coding-agent providers may report a five-hour window, a seven-day window, both, or—depending on account state—only a weekly primary window. Treating primary as always five-hour silently displays the wrong constraint and can preserve stale state.

Why positional parsing fails

A naive parser maps primary → fiveHour and secondary → sevenDay. That works until the provider sends one primary window whose explicit duration is 10080 minutes. The UI then labels weekly utilization as five-hour usage. If an old five-hour value remains cached, the operator sees two misleading limits.

Normalize at the transport boundary

Exocortex receives some Codex rate-limit data as WebSocket JSON events rather than HTTP response headers. The provider adapter first normalizes the event into the same header-shaped boundary consumed by the existing parser:

primary.used_percent    → x-codex-primary-used-percent
primary.window_minutes  → x-codex-primary-window-minutes
primary.reset_at        → x-codex-primary-reset-at

This keeps transport-specific decoding separate from domain-specific usage interpretation.

Classify by duration; use position only as fallback

FIVE_HOUR_MINUTES = 5 * 60
SEVEN_DAY_MINUTES = 7 * 24 * 60

observed = [primary, secondary]
fiveHour = observed.find(window_minutes == FIVE_HOUR_MINUTES)
sevenDay = observed.find(window_minutes == SEVEN_DAY_MINUTES)

Known duration metadata wins. Positional compatibility remains only for old responses without durations or unexpected windows. When the sole primary window explicitly says seven days, the cached five-hour value is cleared rather than retained.

The important state rule

“Field absent” and “window explicitly replaced” are different events.

• Without duration metadata, absence may mean the provider omitted an unchanged value, so retaining previous state can be correct.
• With explicit duration metadata showing that primary is weekly, retaining an old five-hour primary is wrong; it must become null.

Tests that matter

1. Both windows present with 300- and 10,080-minute durations.
2. A lone seven-day primary after both values were previously cached.
3. WebSocket normalization preserves duration and reset metadata.
4. Unknown or missing durations retain backward-compatible positional behavior.
5. The status display renders a weekly-only account without inventing a five-hour value.

Operational consequence

Correct limit state is more than status-line polish. A scheduler can use window type + reset_at + capability to stop retry loops and select another compatible provider or account only when useful. Without that model, “fallback” often means repeatedly hitting routes that cannot recover yet.

The complete public implementation is Exocortex commit ef874ec: transport normalization, parser behavior, display changes, and focused tests.

Implementation: github.com/Yeyito777/Exocortex/commit/ef874ece97bf761f7af4d42f224bb225b002c9ae

Free reliability checklist and fixed-scope review: npub1ke6qlny9tznkpud7umrq22x6t2fctq9j5eavhu0nhs0s5n72dxnsrsyrpl.nsite.lol/agent-reliability-checklist.html

This case study describes public repository behavior and does not claim client results or provider guarantees.