| name | lightflow-video-description |
| description | Generate account-specific video descriptions with a configured model and fail-closed fact validation. |
| version | 0.1.0 |
Video description
Use lightflow.video_description only with the approved edited transcript and
frozen verified facts. The workflow requires LIGHTFLOW_DESCRIPTION_API_URL,
LIGHTFLOW_DESCRIPTION_API_TOKEN, and LIGHTFLOW_DESCRIPTION_MODEL in the
runtime environment. It does not fall back to templates when the model is
unconfigured. Use account_group=zh for the Chinese account and
account_group=overseas for the Russian overseas account.
The model must return JSON with a title, body, and 3–8 hashtags. Any numeric
claim not present in the frozen facts rejects the result for rework.
CLI Usage
lfw run lightflow.video_description \
--input account_group='"zh"' \
--input facts='{"vehicle":"2024 pickup","horsepower":400}' \
--input transcript='"Approved edited transcript text."'
API Usage
curl -sS -X POST http://127.0.0.1:5174/workflows/lightflow.video_description/run \
-H 'content-type: application/json' \
-d '{"inputs":{"account_group":"zh","facts":{"vehicle":"2024 pickup","horsepower":400},"transcript":"Approved edited transcript text."}}'