Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Implement Framer rate limiting, backoff, and idempotency patterns. Use when handling rate limit errors, implementing retry logic, or optimizing API request throughput for Framer. Trigger with phrases like "framer rate limit", "framer throttling", "framer 429", "framer retry", "framer backoff".
.claude/skills/jeremylongshore-framer-rate-limits/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -40% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -10% | 0% |
Handle Framer API rate limits for Server API and plugin operations. The Server API uses WebSocket, so rate limits apply per-connection. CMS operations are limited by collection size and concurrent writes.
| Operation | Limit | Notes | |-----------|-------|-------| | Server API connections | 1 per site | WebSocket, persistent | | CMS setItems | ~100 items/call | Batch larger sets | | CMS getItems | No hard limit | Returns all items | | Plugin API calls | Debounced | Framer throttles internally | | Publish | ~1/minute | Site publishing | | Image upload | Concurrent limit | Via CMS image fields |
typescriptasync function batchSetItems(collection: any, items: any[], batchSize = 100) { for (let i = 0; i < items.length; i += batchSize) { const batch = items.slice(i, i + batchSize); await collection.setItems(batch); console.log(`Synced ${Math.min(i + batchSize, items.length)}/${items.length}`); if (i + batchSize < items.length) { await new Promise(r => setTimeout(r, 1000)); // 1s between batches } } }
typescript// Debounce rapid plugin UI interactions function debounce<T extends (...args: any[]) => any>(fn: T, ms = 300) { let timer: NodeJS.Timeout; return (...args: Parameters<T>) => { clearTimeout(timer); timer = setTimeout(() => fn(...args), ms); }; } const debouncedSync = debounce(async () => { await syncCollection(); }, 500);
typescriptasync function withRetry<T>(fn: () => Promise<T>, maxRetries = 3): Promise<T> { for (let i = 0; i <= maxRetries; i++) { try { return await fn(); } catch (err: any) { if (i === maxRetries) throw err; const delay = 1000 * Math.pow(2, i); console.log(`Retry ${i + 1} in ${delay}ms`); await new Promise(r => setTimeout(r, delay)); } } throw new Error('Unreachable'); }
| Error | Cause | Solution | |-------|-------|----------| | WebSocket disconnected | Connection timeout | Reconnect with backoff | | setItems slow | Large batch | Split into chunks of 100 | | Publish rate limited | Too frequent | Wait 60s between publishes |
For security, see framer-security-basics.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 20,128 | 16,697 | -17% | 1 | 1 | 0% | 3,264 | 3,087 | -5% | 0 | 0 | — |
case-02 | fail→pass | 18,175 | 11,157 | -39% | 1 | 1 | 0% | 3,412 | 3,024 | -11% | 0 | 0 | — |
case-03 | fail→pass | 23,140 | 22,454 | -3% | 1 | 1 | 0% | 3,255 | 4,340 | +33% | 0 | 0 | — |
case-04 | fail→pass | 10,330 | 6,962 | -33% | 1 | 1 | 0% | 1,662 | 991 | -40% | 0 | 0 | — |
case-05 | fail→pass | 16,458 | 10,750 | -35% | 1 | 1 | 0% | 2,032 | 1,864 | -8% | 0 | 0 | — |
case-06 | pass→pass | 12,337 | 8,560 | -31% | 1 | 1 | 0% | 1,922 | 1,330 | -31% | 0 | 0 | — |
case-07 | fail→pass | 17,512 | 6,765 | -61% | 1 | 1 | 0% | 2,093 | 1,879 | -10% | 0 | 0 | — |
case-08 | fail→pass | 14,130 | 8,591 | -39% | 1 | 1 | 0% | 2,510 | 1,356 | -46% | 0 | 0 | — |
case-09 | pass→pass | 15,007 | 10,254 | -32% | 1 | 1 | 0% | 2,342 | 1,790 | -24% | 0 | 0 | — |
case-10 | pass→pass | 15,293 | 8,369 | -45% | 1 | 1 | 0% | 1,577 | 1,246 | -21% | 0 | 0 | — |
case-11 | fail→pass | 11,934 | 2,805 | -76% | 1 | 1 | 0% | 1,100 | 1,188 | +8% | 0 | 0 | — |
case-12 | pass→pass | 24,388 | 13,199 | -46% | 1 | 1 | 0% | 2,963 | 2,687 | -9% | 0 | 0 | — |
case-13 | pass→pass | 19,405 | 3,900 | -80% | 1 | 1 | 0% | 2,218 | 1,468 | -34% | 0 | 0 | — |
case-14 | fail→pass | 11,140 | 7,300 | -34% | 1 | 1 | 0% | 1,904 | 1,183 | -38% | 0 | 0 | — |
case-15 | fail→pass | 12,103 | 17,041 | +41% | 1 | 1 | 0% | 1,876 | 2,969 | +58% | 0 | 0 | — |
case-16 | pass→pass | 13,687 | 10,886 | -20% | 1 | 1 | 0% | 2,741 | 3,016 | +10% | 0 | 0 | — |
case-17 | fail→pass | 32,232 | 6,525 | -80% | 1 | 1 | 0% | 2,045 | 2,094 | +2% | 0 | 0 | — |
case-18 | pass→pass | 14,926 | 2,418 | -84% | 1 | 1 | 0% | 1,492 | 1,111 | -26% | 0 | 0 | — |
case-19 | pass→pass | 13,437 | 12,685 | -6% | 1 | 1 | 0% | 1,794 | 2,458 | +37% | 0 | 0 | — |
case-20 | pass→pass | 18,337 | 16,823 | -8% | 1 | 1 | 0% | 2,099 | 2,397 | +14% | 0 | 0 | — |
case-21 | fail→pass | 8,104 | 3,064 | -62% | 1 | 1 | 0% | 1,494 | 1,390 | -7% | 0 | 0 | — |
case-22 | pass→fail | 9,747 | 9,170 | -6% | 1 | 1 | 0% | 2,030 | 2,763 | +36% | 0 | 0 | — |
case-23 | pass→pass | 11,977 | 7,976 | -33% | 1 | 1 | 0% | 1,572 | 2,658 | +69% | 0 | 0 | — |
case-24 | pass→pass | 16,897 | 11,735 | -31% | 1 | 1 | 0% | 2,462 | 3,194 | +30% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted. The headline lift of +42 percentage points is the difference between those two pass rates over the 24 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.