API reference

Inspect durable operations

List current and completed runs created by conversations, schedules, and automations.

API reference

List runs

Every interactive turn, scheduled job, and automation creates a durable run. Filter the collection by status or trigger, then fetch one run by its id for the full record.

Run ids remain scoped to their workspace. An id from another workspace resolves as not found so callers cannot use identifiers to discover inaccessible work.

status: queued, running, awaiting_approval, done, failed, or cancelled
triggered_by: Optional source identifier for the run
limit: Number of records to return, from 1 through 200
offset: Number of records to skip before the result page

import json
import os
import urllib.request

url = "https://api.trysupervisor.com/workspaces/{workspaceId}/runs?status=running&limit=20&offset=0"
request = urllib.request.Request(url, method="GET")
request.add_header("Authorization", f"Bearer {os.environ['SUPERVISOR_API_KEY']}")
request.add_header("Content-Type", "application/json")

with urllib.request.urlopen(request) as response:
    result = json.load(response)
    print(json.dumps(result, indent=2))

Read one run

Request GET /workspaces/{workspaceId}/runs/{runId} after selecting an item from the collection. Use the returned status and timestamps to update progress in your own interface.