openclaw - 💡(How to fix) Fix Cron tasks fail due to rate-limit cooldown timeout in 5.20+

Problem

After upgrading to OpenClaw 5.20+, cron tasks are failing with rate-limit related errors. The issue appears to be a regression in how the rate-limit cooldown is handled for cron-triggered tasks.

Analysis

1. 5.20 introduced more precise `getSoonestCooldownExpiry` auth profile tracking
2. When a cron triggers, it detects the credential is in cooldown, logs `"Rate-limited — ready in ~7s"` and waits
3. However, cron task timeout may be too short — after the 7s cooldown wait, the task may already be considered timed-out or failed
4. 5.22 lazy-load optimization may worsen this: slower gateway startup → delayed channel handshake → cron tasks arrive later, compounding the timeout issue

Expected Behavior

Cron tasks should either:

• Be queued until the cooldown expires, then execute normally
• Respect a configurable timeout that accounts for cooldown waiting
• Not silently fail without clear error messaging

Version

• OpenClaw: 5.20+ (confirmed in 5.20, 5.22)
• Affects: All cron jobs that use rate-limited API credentials (e.g., Feishu API, model providers)

Logs

Typical error pattern: ``` [scheduled] job=xxx triggered Rate-limited — ready in ~7s (credential cooldown detected, waiting) ... (7s later) Error: task timed out OR task failed ```

Suggested Fix

Options:

1. Increase default cron timeout to account for cooldown duration
2. Pre-warm credentials before cron execution to avoid cooldown
3. Add retry logic with cooldown-aware backoff for cron tasks
4. Log clearer warnings when a task is skipped due to cooldown timeout