Skip to content
QuantCDN

Create Durable Orchestration

POST
/api/v3/organizations/{organisation}/ai/orchestrations
curl --request POST \
  --url https://dashboard.quantcdn.io/api/v3/organizations/example/ai/orchestrations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{ "name": "Process batch documents", "description": "example", "agentId": "example", "toolId": "example", "workflowId": "example", "inputSource": { "type": "static", "items": [ {} ], "taskQuery": {}, "generatorPrompt": "example" }, "batchSize": 10, "concurrency": 1, "stopCondition": { "type": "all_complete", "maxIterations": 1, "conditionPrompt": "example" }, "assignedSkills": [ "example" ], "context": {}, "autoStart": true }'

Create a new durable orchestration for batch processing. * * Input Sources: * - static: Process a fixed list of items * - task_query: Process tasks matching a query * - generator: AI generates items from a prompt * * Stop Conditions: * - all_complete: Stop when all items processed * - max_iterations: Stop after N iterations * - condition: AI evaluates a prompt to decide * - manual: Run until manually stopped * * Auto-start: * By default, the orchestration starts immediately. Set autoStart: false to create in pending state.

Authorizations

Section titled “Authorizations”

Parameters

Section titled “Parameters”

Path Parameters

Section titled “Path Parameters”
organisation
required
string

The organisation machine name

Request Bodyrequired

Section titled “Request Bodyrequired”
Media typeapplication/json
object
name
required

Orchestration name

string
Example
Process batch documents
description

Optional description

string
nullable
agentId

Agent to process items

string
nullable
toolId

Tool to execute for items

string
nullable
workflowId

Workflow to run for items

string
nullable
inputSource
required
object
type
required

Input source type (api type not yet supported)

string
Allowed values: static task_query generator
items

Static items (for type=static)

Array<object>
object
taskQuery

Task query filters (for type=task_query)

object
generatorPrompt

AI prompt (for type=generator)

string
batchSize

Items per batch

integer
default: 10 >= 1 <= 100
concurrency

Concurrent items within a batch

integer
default: 1 >= 1 <= 10
stopCondition
object
type
string
default: all_complete
Allowed values: all_complete max_iterations condition manual
maxIterations

Max iterations (for type=max_iterations)

integer
>= 1
conditionPrompt

AI prompt to evaluate stop (for type=condition)

string
assignedSkills

Skill IDs to assign

Array<string>
context

Optional context merged into each item before processing (e.g. shared credentials or endpoints)

object
autoStart

Whether to start immediately

boolean
default: true

Responses

Section titled “Responses”

201

Section titled “201”

Orchestration created

Media typeapplication/json
object
Examplegenerated
{}

400

Section titled “400”

Invalid request

403

Section titled “403”

Access denied

500

Section titled “500”

Failed to create orchestration