# Request an endpoint

Need data FetchAPI doesn't serve yet? Ask. We find the cheapest working source, check its licence and price, and answer.
When we add it, the answer names the endpoint to call.

```
POST https://fetchapi.co/api/requests
content-type: application/json
x-api-key: gw_...            (optional)

{"want": "domain age: input a domain, output its registration date",
 "input_example": "example.com", "output_example": "1995-08-14",
 "use_case": "score new sign-ups", "volume": "5,000 a month", "budget": "$0.001 a call"}
```

Answer: `{"status":"success","data":{"id":42,"status":"new","status_url":"https://fetchapi.co/api/requests/42?token=...",...}}`.
Keep `status_url`: it shows our answer (`GET`, no key needed). We don't send email; `email` is an optional contact field.

Read the answers to your requests (your key): `GET https://fetchapi.co/api/requests` → each request's `status`
(`new`, `investigating`, `live`, `declined`, `duplicate`), our `reply` (source, price, licence) and `reply_endpoint`
(what to call now). MCP: tool `request_endpoint {want, input_example?, output_example?, use_case?, volume?, budget?}` (returns the same status_url).
Form: https://fetchapi.co/catalog#request. Limit: 10 requests an hour per network.
