> ## Documentation Index
> Fetch the complete documentation index at: https://docs.skilldiscs.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Rate Limits

> How requests are throttled and how to back off.

## Limits

| Limit                   | Value                                  |
| ----------------------- | -------------------------------------- |
| Requests / minute / key | **100**                                |
| Batch size              | **10 ids** per `/disks/batch` call     |
| Search query length     | First **1000 characters** are embedded |
| Search result count     | Up to **25** sections                  |
| List page size          | Up to **100** disks                    |

## Detecting throttling

A throttled request returns HTTP `429` with body:

```json theme={null}
{
  "error": "rate_limit_exceeded",
  "retry_after": 42
}
```

`retry_after` is in **seconds**. Wait that long, then retry.

## Backoff strategy

```javascript theme={null}
async function fetchWithBackoff(url, opts, attempt = 0) {
  const r = await fetch(url, opts)
  if (r.status !== 429) return r
  const { retry_after = 60 } = await r.json().catch(() => ({}))
  if (attempt >= 3) throw new Error('rate_limited')
  const wait = (retry_after + Math.random() * 5) * 1000
  await new Promise(res => setTimeout(res, wait))
  return fetchWithBackoff(url, opts, attempt + 1)
}
```

Add jitter to avoid thundering-herd retries from parallel agents sharing one key.

## Need higher limits?

Mention your use case at [skilldiscs.com/settings](https://skilldiscs.com/settings) — limits are per-account and adjustable on request for production agents.