Quickstart

This page shows how to use OpenAPI for a basic content-creation flow.

The examples below use slides_video for a basic flow. The actual types, ratios, page counts, qualities, languages, voices, and styles still need to come from /catalog.

Get the API Key Before You Start

OpenAPI uses the user's own API key.

If you do not have one yet, go to the main site's API Key page:

activate the API key
copy the API key
come back here and validate it with GET /me

If you are integrating for end users, this step should be part of your connect flow.

Prepare

export KGP_BASE_URL="https://api.kpainter.ai/openapi/v1"export KGP_API_KEY="<your_api_key>"

import requests BASE_URL = "https://api.kpainter.ai/openapi/v1"HEADERS = {    "X-KGP-Api-Key": "<your_api_key>",}

Step 1: Validate the Key

Once you have the key, validate it before creating anything:

curl -s "$KGP_BASE_URL/me" \  -H "X-KGP-Api-Key: $KGP_API_KEY"

me_response = requests.get(f"{BASE_URL}/me", headers=HEADERS, timeout=30)me_response.raise_for_status()print(me_response.json())

At minimum, confirm:

the key is valid
it is a user-level key that can create content

Step 2: Read the Catalog

curl -s "$KGP_BASE_URL/catalog" \  -H "X-KGP-Api-Key: $KGP_API_KEY"

catalog_response = requests.get(f"{BASE_URL}/catalog", headers=HEADERS, timeout=30)catalog_response.raise_for_status()catalog = catalog_response.json()print(catalog["content_types"])

Use it to load:

supported content types
per-type aspect ratio, quality, duration, and page limits
languages, voices, and styles

Step 3: Create Content

curl -s "$KGP_BASE_URL/creations" \  -H "Content-Type: application/json" \  -H "X-KGP-Api-Key: $KGP_API_KEY" \  -d '{    "type": "slides_video",    "prompt": "Explain how transformer attention works in 6 pages",    "language": "en",    "aspect_ratio": "16:9",    "scene_count": 6  }'

create_payload = {    "type": "slides_video",    "prompt": "Explain how transformer attention works in 6 pages",    "language": "en",    "aspect_ratio": "16:9",    "scene_count": 6,} create_response = requests.post(    f"{BASE_URL}/creations",    headers={**HEADERS, "Content-Type": "application/json"},    json=create_payload,    timeout=30,)create_response.raise_for_status() creation = create_response.json()creation_id = creation["creation_id"]job_id = creation["job"]["job_id"] print({"creation_id": creation_id, "job_id": job_id})

A basic request usually contains:

type
prompt
the required type-specific fields

Step 4: Poll the Job

curl -s "$KGP_BASE_URL/creations/jobs/<job_id>" \  -H "X-KGP-Api-Key: $KGP_API_KEY"

import time while True:    job_response = requests.get(        f"{BASE_URL}/creations/jobs/{job_id}",        headers=HEADERS,        timeout=30,    )    job_response.raise_for_status()    job = job_response.json()     print(job["status"], job.get("progress_pct"), job.get("current_step"))     if job["status"] == "succeeded":        break    if job["status"] in {"failed", "cancelled"}:        raise RuntimeError(job)     time.sleep(3)

When the job succeeds, fetch the detail response.

Step 5: Read the Detail

curl -s "$KGP_BASE_URL/creations/<creation_id>" \  -H "X-KGP-Api-Key: $KGP_API_KEY"

detail_response = requests.get(    f"{BASE_URL}/creations/{creation_id}",    headers=HEADERS,    timeout=30,)detail_response.raise_for_status()detail = detail_response.json() print(detail["main_url"])print(detail.get("artifacts", []))print(detail.get("scenes", []))

This is where you read:

main_url
artifacts[]
scenes[]

Endpoints Used In This Flow

GET /me
GET /catalog
POST /creations
GET /creations/jobs/{job_id}
GET /creations/{creation_id}