Overview

Clawdcat is a small, private API gateway. It exposes an HTTP interface compatible with common chat-completion clients, so existing SDKs work by changing only the base URL and key.

Base URL:

https://api.clawdcat.io/v1

All requests and responses use JSON over HTTPS. Plain HTTP is redirected. Access is by invitation; keys are issued per user and can be revoked at any time.

Authentication

Send your key in the Authorization header as a bearer token.

curl https://api.clawdcat.io/v1/models \
  -H "Authorization: Bearer $CLAWDCAT_API_KEY"

Keys start with sk-. Treat them like passwords: do not embed them in client-side code or commit them to source control.

Endpoints

Method	Path	Description
GET	`/v1/models`	List models available to your key.
POST	`/v1/chat/completions`	Create a chat completion.
POST	`/v1/responses`	Create a response (newer request format).
GET	`/health`	Service health. Returns `{"status":"ok"}`.

Example:

curl https://api.clawdcat.io/v1/chat/completions \
  -H "Authorization: Bearer $CLAWDCAT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "default",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Streaming

Set "stream": true to receive server-sent events. Each event carries a partial delta; the stream ends with data: [DONE]. Idle connections are closed after 60 minutes.

Rate limits

Limits are applied per key. When a limit is reached the API returns 429 with a Retry-After header in seconds. Back off and retry; do not retry in a tight loop.

Limit	Default
Concurrent requests	4
Requests per minute	60
Maximum request body	256 MB

Errors

Status	Meaning
`400`	The request body is malformed or a parameter is invalid.
`401`	Missing or invalid API key.
`403`	The key is not allowed to use this model.
`429`	Rate limit reached. See `Retry-After`.
`502` / `503`	Upstream temporarily unavailable. Retry with backoff.

Error bodies have the form {"error": {"type": "...", "message": "..."}}.

Changelog

Date	Change
2026-10	Added `/v1/responses`. Longer streaming timeout.
2026-08	Per-key concurrency limits.
2026-07	Initial release.