Bellwork
Start free
OrgSequencesSequenceidEnrollmentsEnrollmentidMessagesMessageid

Edit a planned message

Replace the model's words with a person's. The runner consumes stored messages verbatim, so this edit IS what sends when the step comes due.

PATCH
/org/sequences/{sequenceId}/enrollments/{enrollmentId}/messages/{messageId}

Replace the model's words with a person's. The runner consumes stored messages verbatim, so this edit IS what sends when the step comes due.

Authorization

AuthorizationBearer <token>

API key (sk_live_... prefix). Generate keys in the UI under Settings > API Keys, then send it as Authorization: Bearer sk_live_...

In: header

Path Parameters

sequenceId*string
Formatuuid
enrollmentId*string
Formatuuid
messageId*string
Formatuuid

Request Body

application/json

subject?string
Lengthlength <= 500
body?string
Lengthlength <= 20000

Response Body

application/json

application/json

application/json

application/json

application/json

curl -X PATCH "https://loading/api/v1/org/sequences/497f6eca-6276-4993-bfeb-53cbbbba6f08/enrollments/497f6eca-6276-4993-bfeb-53cbbbba6f08/messages/497f6eca-6276-4993-bfeb-53cbbbba6f08" \  -H "Content-Type: application/json" \  -d '{}'
{
  "data": {
    "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
    "node_id": "6e0bdcac-9717-4a6f-972e-97489dffeb0c",
    "node_label": "string",
    "node_kind": "string",
    "subject": "string",
    "body": "string",
    "is_pending": true
  }
}
{
  "error": "string",
  "message": "string",
  "statusCode": -9007199254740991
}
{
  "error": "string",
  "message": "string",
  "statusCode": -9007199254740991
}
{
  "error": "module_required",
  "module": "string",
  "message": "string",
  "contact": "string"
}
{
  "error": "string",
  "message": "string",
  "statusCode": -9007199254740991
}

One person's planned messages GET

The whole conversation-to-be, in the order the sequence will happen to this person. Written at enrollment; whatever is stored here is what each step sends verbatim when it comes due. The message currently parked in the review queue is flagged `is_pending` — send or skip it there.

Write a planned message again POST

Ask the writer for another version of one message, optionally saying what should change. Unlike the PATCH beside it, this spends a model call — and it is the only way to get a new version that still knows the district's research, the org's note on the contact, and what the earlier steps in the chain already said. Replaces the message in place, so a message parked for review stays parked. On failure the existing message is left untouched.