Skip to main content
PATCH
Update the workspace AI defaults

Authorizations

Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

Partial update: only the fields present in the request are written.

default_language
string | null
default_reply_mode
enum<string> | null

How far the AI reply layer may go on this sequence.

Auto-drafting is PHASE2 §8; the self-sending auto-pilot is INB-15. OFF is the default everywhere: a workspace opts in. AUTO_SEND is a further EXPLICIT opt-in: a HIGH-confidence draft on a healthy thread is approved automatically — through the exact same approve path a human uses, so sending stays one code path — while anything uncertain (lower confidence, a Negative/Wrong-contact label, OOO, a bounce, or a thread the agent has already answered five times) waits in Needs approval exactly like DRAFT_AND_QUEUE. An auto-sent draft records NO approver: a NULL approved_by_user_id on a SENT draft is the honest audit trail.

Available options:
off,
draft_and_queue,
auto_send
default_tone
enum<string> | null

How a generated reply should read.

Available options:
professional,
friendly,
direct,
casual
labelling_enabled
boolean | null

Response

Successful Response

The workspace's AI defaults — INB-03's auto-label switch lives here.

default_language
string | null
required
default_reply_mode
enum<string>
required

How far the AI reply layer may go on this sequence.

Auto-drafting is PHASE2 §8; the self-sending auto-pilot is INB-15. OFF is the default everywhere: a workspace opts in. AUTO_SEND is a further EXPLICIT opt-in: a HIGH-confidence draft on a healthy thread is approved automatically — through the exact same approve path a human uses, so sending stays one code path — while anything uncertain (lower confidence, a Negative/Wrong-contact label, OOO, a bounce, or a thread the agent has already answered five times) waits in Needs approval exactly like DRAFT_AND_QUEUE. An auto-sent draft records NO approver: a NULL approved_by_user_id on a SENT draft is the honest audit trail.

Available options:
off,
draft_and_queue,
auto_send
default_tone
enum<string>
required

How a generated reply should read.

Available options:
professional,
friendly,
direct,
casual
labelling_enabled
boolean
required