Smithery is now a part of Arcade.dev! Read more in our announcement here
Execute pre-launch production readiness checklist for Kling AI. Use when preparing to deploy video generation to production...
Checklist covering authentication, error handling, cost controls, monitoring, security, and content policy before deploying Kling AI video generation to production.
.env in repo)Authorization: Bearer <token> format verifiedtask_status: "failed" logs task_status_msgduration sent as string "5" not integer 5standard mode used for non-final renders# Pre-batch credit check
credits_needed = len(prompts) * 10 # 10 credits per 5s standard
if credits_needed > DAILY_BUDGET:
raise RuntimeError(f"Batch needs {credits_needed}, budget is {DAILY_BUDGET}")
callback_url used instead of polling in productionrequests.Session()# Connection pooling
session = requests.Session()
adapter = requests.adapters.HTTPAdapter(pool_connections=5, pool_maxsize=10)
session.mount("https://", adapter)
from kling_client import KlingClient
c = KlingClient()
result = c.text_to_video("test: blue sky with clouds", duration=5, mode="standard")
assert result["videos"][0]["url"], "No video URL"
print("READY FOR PRODUCTION")
Produce a production receipt with release ID, environment, smoke-test brief classification, aggregate health/task result, policy/rights/budget checks, draft destination, approver, retention/removal proof, and rollback reference. Exclude prompts, asset URLs, and credentials.
release=r31; env=staging; brief=synthetic-smoke; policy=pass; rights=pass; destination=draft-only; approval=pending; rollback=r30 supports canary approval.
