FruxonDocs
ApiAgent

Streams AI-assisted agent building instructions via Server-Sent Events

Drives the in-Studio Copilot. The response is delivered incrementally over an SSE stream rather than as a single JSON body, so clients must read the event stream to completion. Prior turns for the same agent are replayed as context, enabling multi-turn refinement of the agent's configuration.

POST
/v1/tenants/{tenant}/agents/{agent}/copilot:chat

Authorization

Bearer
AuthorizationBearer <token>

JWT Authorization header using the Bearer scheme. Enter 'Bearer' [space] and then your token.

In: header

Path Parameters

agent*string

The agent identifier.

tenant*string

The tenant identifier

Request Body

The copilot chat request containing the user message and context.

message?string|null

The user's conversational message (e.g. "Add a step that sends a Slack message").

conversationId?string|null

Optional conversation ID to continue an existing conversation. When null, a new conversation is created.

Formatuuid
canvasState?

Snapshot of the workflow canvas sent with each Copilot request.

availableTools?|null

Platform tools available for the agent to use.

availableProviders?|null

AI/LLM providers available on the platform.

agentType?string|null

The type of agent being built (e.g. "chat").

[key: string]?never

Response Body

curl -X POST "https://api.fruxon.com/v1/tenants/string/agents/string/copilot:chat" \  -H "Content-Type: application/json" \  -d '{}'
"string"
{
  "type": "string",
  "title": "string",
  "status": 0,
  "detail": "string",
  "instance": "string",
  "property1": null,
  "property2": null
}
Empty
Empty
{
  "type": "string",
  "title": "string",
  "status": 0,
  "detail": "string",
  "instance": "string",
  "property1": null,
  "property2": null
}