Script API#

Read scripts, override them per tenant, and analyze them. All endpoints are tenant-bound (X-Tenant).

Read scripts#

GET/v2/audio/scripts

Returns the effective script set: default scripts overlaid with tenant overrides. With ?analyze=1, every entry comes with its full analysis.

bash

curl -s -H "X-Tenant: bbe" \
  "https://api.callfluence.staging.bestbrandseverywhere.com/v2/audio/scripts?analyze=1"

Every entry carries its origin: "source": "default" or "source": "tenant".

Single script with analysis#

GET/v2/audio/scripts/{touchpoint}

Override a script#

PUT/v2/audio/scripts/{touchpoint}

bash

curl -s -X PUT -H "Content-Type: application/json" -H "X-Tenant: bbe" \
  -d '{"text": "Einen Moment bitte, Ihr Anliegen wird sofort weiterbearbeitet."}' \
  https://api.callfluence.staging.bestbrandseverywhere.com/v2/audio/scripts/call.hold

The response includes the immediate analysis of the new text — rule violations surface at save time, not in production:

json

{
  "touchpoint_id": "call.hold",
  "text": "Einen Moment bitte, Ihr Anliegen wird sofort weiterbearbeitet.",
  "analysis": { "words": 8, "durationSec": 4, "variables": [], "findings": [] }
}

Reset to default#

DELETE/v2/audio/scripts/{touchpoint}

Removes the tenant override; the default script applies again.

What the analysis checks#

Rule	Level	Check
`dauer`	warning	Estimated speaking time (110 words/minute) against the touchpoint's `maxSeconds`
`kundensicht`	warning	Vendor perspective (we/us/our) instead of addressing the customer
`umlaute`	error	ASCII substitutions (ue/ae/oe) in spoken text
`einwortzeile`	warning	One-word lines addressed at the customer
`doppelte-kennzeichnung`	warning	AI disclosure inside the script that is already prepended automatically

Audio settings#

Recording and transcription are customer switches per tenant:

GET/v2/audio/settings

PATCH/v2/audio/settings

bash

curl -s -X PATCH -H "Content-Type: application/json" -H "X-Tenant: bbe" \
  -d '{"recordingEnabled": false}' \
  https://api.callfluence.staging.bestbrandseverywhere.com/v2/audio/settings

json

{ "tenant": "bbe", "recordingEnabled": false, "transcriptionEnabled": true }