Rate Limiting

To keep the service stable for everyone, the Gearment API enforces a rate limit per team.

Setting	Value
Request limit	100 requests per 10 seconds
Scope	Counted per Gearment team, across every credential and every endpoint
Penalty	Once the limit is hit, further requests are blocked for 1 minute
Retry signal	The Retry-After response header states how long the block lasts

The window slides: the limiter counts the requests your team made in the previous 10 seconds. Because the budget belongs to the team, parallel workers that share a team share the same budget.

When you exceed the limit

Throttled requests return 429 Too Many Requests:

HTTP/1.1 429 Too Many Requests
Retry-After: 60s
Content-Type: application/json

{
  "status": "error",
  "message": "rate limit exceeded - retry after 60 seconds",
  "request_id": "b7f42c7f-a1c8-4241-8a99-cc1ece6b0351"
}

The block is not lifted early by waiting out the 10-second window - once you are blocked, the Retry-After duration is what applies.

Best practices

• Retry 429 responses with exponential backoff, and honour Retry-After instead of retrying immediately.
• Monitor your API usage so a runaway worker does not spend the team's budget.
• Batch updates where the API supports it, and avoid polling for state you can receive as a webhook.