Skip to main content

​
Kling AI

Shanone’s Kling AI integration gives your agent 6 tools for AI video generation — creating video from text or an image, extending existing clips, syncing lip movement to audio, and tracking generation tasks through to completion.

​
Getting Started

1

Get your Kling AI Access Key and Secret Key

Generate an API key pair from your Kling AI account.
2

Add them to Shanone

Open Integrations in the Shanone dashboard, select Kling AI, and enter your Access Key and Secret Key.
3

Retry your request

Once saved, shanone_execute_tool calls for kling_* tools will succeed.

​
Available Tools

Shanone provides 6 tools for Kling AI, organized into these categories:
kling_text_to_video, kling_image_to_video — generate a video clip from a text prompt or a starting image
kling_extend_video, kling_lip_sync — extend a previously generated clip, or sync a character’s lip movement to an audio track
kling_get_task, kling_wait_for_task — check a generation task’s status once, or block until it completes (or times out) and return the resulting video URL

​
Common Use Cases

Text-to-video generation

Turn a scene description into a short AI-generated video clip

Image animation

Animate a still image into a moving video clip

Dubbed lip-sync

Sync a character’s mouth movement to a new audio track for localization

Extended sequences

Chain kling_extend_video calls to build a longer sequence from an initial clip

​
Troubleshooting

Video generation can take longer than the default wait window — increase max_wait_seconds, or fall back to polling kling_get_task yourself on a longer interval rather than blocking in one call.
Check the error field returned by kling_get_task — common causes are prompts that violate content policy or a source image that doesn’t meet Kling’s resolution/format requirements.
Lip-sync quality depends on clean, isolated speech audio in the input track — background noise or overlapping speakers will degrade sync accuracy.