API Rate Limits

Limits

• Per IP address: 60 requests per minute. This applies to API requests with or without an API key.
• Per API key: 100 requests per minute. A request with a key counts toward both limits, so a single client on one IP address gets about 60 requests per minute.
• MCP: Each tool call is one request to the MCP server plus one or two REST API calls made with your key. Each of these counts toward your key's limit.
• ChangeForge widget feed: This has its own limit instead, 600 requests per minute per IP address, because many of your visitors can share one address.

Limits are approximate. Each window lasts 1 minute.

Rate limit headers

API responses don't include usage headers, so keep track of your own request rate. The only rate limit header is Retry-After, which comes with some 429 responses.

What happens when you hit the limit

You get a 429 Too Many Requests response.

• Per-key limit: The response has a Retry-After header with the number of seconds to wait, and a JSON body like this:

  { "error": "rate_limited", "message": "Rate limit exceeded", "retry_after": 42 }

• Per-IP limit: The body is the plain text Too many requests, with no Retry-After header. Wait up to 1 minute, then retry.

Subscriber endpoints

Public subscribe requests have a separate limit: 10 new subscriptions per account per minute. This covers status page and changelog subscriptions. Extra requests get a 429.