Added

OnSched API 3.25.0 — V1 connect/token on v3

V1 token drop-in

POST /connect/token (and POST /identity/connect/token) on the v3 host is a drop-in for v1 Identity POST /connect/token.

Send application/x-www-form-urlencoded client_id, client_secret, scope=OnSchedApi, and grant_type=client_credentials. HTTP Basic is also accepted. Many production client_id values start with live and sandbox values with sbox; those prefixes are hints only. Leftover v1 Identity tokens and new /connect/token Machine Tokens can be used on the same host until Identity is turned off.

The response is { access_token, expires_in: 3600, token_type: "Bearer", scope: "OnSchedApi" }. Treat access_token as opaque. Use it as Authorization: Bearer on /consumer/v1 and /setup/v1.

Production token URL: https://v3.onsched.com/connect/token. Sandbox token URL: https://api-stage.onsched.com/connect/token.

Native POST /v3/oauth/token is unchanged and still rejects OnSchedApi. Those native tokens are not accepted on /consumer/v1 or /setup/v1.

A Machine Token can do only what its own scopes allow. read without write is limited to pure reads. Creates, updates, deletes, and reads that would create or restore records return 403. POST /connect/token with OnSchedApi still issues read write.

V1 appointment reschedule

PUT /consumer/v1/appointments/:id/reschedule now updates legacy ID mapping when v3 marks the original appointment RE and creates a successor. Later cancel, book, confirm, or GET calls that reuse the same v1 appointment id reach the live booking instead of the rescheduled-away husk.