> For the complete documentation index, see [llms.txt](https://docs.dapplooker.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.dapplooker.com/resources/best-practices.md).

# Best Practices

## 1. Use Caching Where Data Does Not Need Real-Time Refresh

Cache responses when second-by-second updates are unnecessary.

Recommended for:

* Token metadata
* Historical market data
* Token directory queries
* Slow-changing holder metrics

Benefits:

* Lower request volume
* Faster app performance
* Reduced costs
* Better rate-limit efficiency

***

## 2. Batch Requests for Multiple Assets

When querying multiple tokens, send batched requests where supported.

Use comma-separated:

* token\_ids
* token\_tickers
* token\_addresses

Recommended batch size:

```
Up to 30 assets per request
```

Benefits:

* Lower network overhead
* Faster scans
* Cleaner workflows

***

## 3. Use Real-Time Polling Selectively

Poll high-frequency endpoints only where freshness matters.

Examples:

* Perp Intelligence
* Trending Tokens
* Smart Money Trends

Use slower refresh intervals for:

* Token Directory
* Historical Data
* Staking Metrics

***

## 4. Handle Errors Gracefully

Always inspect HTTP status codes and implement retry logic.

Common responses:

| Code | Meaning                           | Recommended Action             |
| ---- | --------------------------------- | ------------------------------ |
| 400  | Bad request                       | Validate parameters            |
| 401  | Unauthorized                      | Check API key / auth           |
| 402  | Payment required                  | Complete x402 payment flow     |
| 423  | Temporarily locked / rate limited | Retry with backoff             |
| 429  | Too many requests                 | Slow down request rate         |
| 500+ | Server issue                      | Retry with exponential backoff |

***

## 5. Implement Retries with Backoff

For temporary failures:

* exponential backoff
* jittered retries
* capped retry count

Avoid aggressive retry loops.

***

## 6. Use the Right Access Model

### API Keys

Best for:

* dashboards
* apps
* predictable workloads

### x402 Payments

Best for:

* AI agents
* bots
* dynamic usage
* pay-per-call workloads

***

## 7. Separate Core Data from Intelligence Calls

Use lower-frequency refresh for:

* Token Directory
* Historical Market Data

Use higher-frequency refresh for:

* Perp Intelligence
* Smart Money Trends
* Trending Tokens

This improves efficiency.

***

## 8. Design for Schema Stability

Use typed parsers and version-safe wrappers.

Do not hardcode assumptions beyond documented fields.

***

## 9. Monitor Usage and Limits

Track:

* request volume
* latency
* error rates
* payment failures
* rate-limit events

Reach out to DappLooker for higher limits or enterprise throughput.

***

## 10. Stay Updated

Check docs regularly for:

* new endpoints
* schema upgrades
* additional supported chains
* new DEX integrations
* new agent tools

***

## Pro Tip for AI Agents

Use direct APIs for deterministic outputs and MCP / NLP APIs for flexible reasoning workflows.