kling-video · acedatacloud — Claude Code skill: what it does, how to install
kling-video generates AI video via the AceDataCloud Kuaishou Kling API. Use it to render text prompts, animate still frames, extend existing clips, and run lip sync.
Apache-2.0 240 installs ★ 17 updated 2026-10-25
npx skills add AceDataCloud/Skills --skill kling-video -a claude-codeWhat kling-video does
The skill dispatches requests to the AceDataCloud platform, covering Kuaishou Kling models from kling-v1 to kling-v3, kling-v3-omni, and kling-o1. It handles text-to-video, animates static images using start and end frames, extends previously rendered clips by ID, and drives lip-sync workflows via /kling/lip-sync.
Output quality relies on std, pro, or native 4k tiers. Native 4K works exclusively on kling-v3 and kling-v3-omni, and it conflicts with camera_control.
Calls require an active key stored in ACEDATACLOUD_API_TOKEN inside ~/.openclaw/.env or exported in your shell. The API rejects requests without it. Video rendering runs asynchronously, so follow-up queries poll /kling/tasks until the task finishes.
What it helps with
- generate video clips from text prompts using the
kling-v3model - animate static images with optional start and end frames
- extend completed videos by ID with additional seconds
- lip-sync a character in a source video to an audio track via
/kling/lip-sync
Example prompts for the agent
- Generate a 5-second video of a futuristic city using kling-v3 in pro mode
- Animate scene.jpg with end-scene.jpg as the closing frame
- Extend the clip with ID video_123 by pulling the camera back
How to install kling-video
1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill kling-video -a claude-code2Manually: put the files in place
~/.claude/skills/kling-video/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/kling-video/
1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill kling-video -a codex2Manually: put the files in place
.agents/skills/kling-video/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill kling-video -a cursor2Manually: put the files in place
.cursor/skills/kling-video/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install AceDataCloud/Skills skills/kling-video2Manually: put the files in place
.github/skills/kling-video/SKILL.md1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill kling-video -a gemini-cli2Platform CLI
gemini skills install https://github.com/AceDataCloud/Skills --path skills/kling-videoManually: put the files in place
.devin/skills/kling-video/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill kling-video -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about kling-video
How to install kling-video in Claude Code?
Run npx skills add AceDataCloud/Skills --skill kling-video -a claude-code.
What does the kling-video skill do?
It interfaces with the AceDataCloud API to render Kling AI videos from text or images, extend existing footage, and apply lip sync.
What credentials are required to use this skill?
You need an AceDataCloud API key assigned to ACEDATACLOUD_API_TOKEN in ~/.openclaw/.env or exported directly into your environment.
Source text
SKILL.md · Apache-2.0
---
name: kling-video
description: Create Kling videos and commercial media through AceDataCloud. Use for text/image video generation, Turbo, storyboards, motion control, talking photos, lip sync, product studio, commercial voiceovers, and supported reference inputs.
license: Apache-2.0
metadata:
author: acedatacloud
version: "1.0"
compatibility: Requires ACEDATACLOUD_API_TOKEN in .env file (get an API key at https://platform.acedata.cloud/console/applications).
---
# Kling Video Generation
Generate AI videos through AceDataCloud's Kuaishou Kling API.
## First-time setup
1. Install the official Skill with `clawhub install @acedatacloud/kling`. Check that the ClawHub publisher is **@acedatacloud**.
2. Sign in at [AceDataCloud → Applications](https://platform.acedata.cloud/console/applications), open **General application**, and check Kling access, balance, and current pricing. Copy its API key, or select **Manage Keys → Create** for a separate key. Copy the key value without `Bearer `; a `platform-...` management token is not a service API key.
3. If you enable **Allowed APIs** restrictions, include `/kling/videos` and `/kling/tasks`. Advanced workflows below may call additional routes.
4. On the OpenClaw Gateway host, put `ACEDATACLOUD_API_TOKEN=your_api_key_here` in `~/.openclaw/.env`, then restart the Gateway and start a new session. For a direct terminal test, run `export ACEDATACLOUD_API_TOKEN="your_api_key_here"` in that shell. Never put the key in a prompt or commit an `.env` file.
5. Run the **Quick Start** below. If this workflow returns a task ID, query that same task until it finishes; do not submit the same billable request again just to check progress.
[Illustrated key setup, first run, and troubleshooting](https://github.com/AceDataCloud/Skills#first-run-with-clawhub-and-openclaw) · [Token details](https://github.com/AceDataCloud/Skills/blob/main/skills/_shared/authentication.md)
## Quick Start
```bash
curl -X POST https://api.acedata.cloud/kling/videos \
-H "Authorization: Bearer $ACEDATACLOUD_API_TOKEN" \
-H "Content-Type: application/json" \
-d '{"action": "text2video", "prompt": "a cat playing piano on a rooftop at sunset", "model": "kling-v3", "mode": "std", "duration": 5}'
```
> **Async:** See [async task polling](https://github.com/AceDataCloud/Skills/blob/main/skills/_shared/async-tasks.md). Poll via `POST /kling/tasks` with `{"id": "..."}`.
## Models
| Model | Quality | Best For |
|-------|---------|----------|
| `kling-v3-turbo` | Fast | 720p/1080p, integer 3–15 seconds, included native audio with no off switch |
| `kling-v3` | Latest | Best quality, flexible 3–15s duration, optional audio generation |
| `kling-v3-omni` | Latest | V3 Omni model with audio plus image/video references, flexible 3–15s duration |
| `kling-v2-6` | High | High-quality output with optional audio (pro mode) |
| `kling-v2-5-turbo` | High + Fast | Best speed/quality trade-off |
| `kling-v2-master` | High | High-quality output |
| `kling-v2-1-master` | High | Improved v2 |
| `kling-v1-6` | Improved | Better quality than v1 |
| `kling-v1` | Standard | Basic generation, lowest cost |
| `kling-o1` | Premium | Independent O1 model with image/video references, 5s only |
## Quality Modes
| Mode | Speed | Cost | Use For |
|------|-------|------|---------|
| `std` (Standard) | Slower | Lower | Draft/preview |
| `pro` (Professional) | Faster | Higher | Final output |
| `4k` (Native 4K) | — | Premium | Native 4K output — only `kling-v3` and `kling-v3-omni`; incompatible with `camera_control` |
## Workflows
### 1. Text-to-Video
```json
POST /kling/videos
{
"action": "text2video",
"prompt": "a futuristic city with flying cars",
"model": "kling-v3",
"mode": "std",
"duration": 5,
"aspect_ratio": "16:9"
}
```
### 2. Image-to-Video
Animate a still image. Optionally specify an ending frame.
```json
POST /kling/videos
{
"action": "image2video",
"prompt": "the scene slowly comes alive with movement",
"start_image_url": "https://example.com/scene.jpg",
"end_image_url": "https://example.com/end-scene.jpg",
"model": "kling-v3",
"mode": "pro"
}
```
### 3. Omni References
Use `kling-o1` or `kling-v3-omni` with reference images and/or one reference video. Cite each item in the prompt using its one-based token.
```json
POST /kling/videos
{
"action": "text2video",
"prompt": "turn <<<video_1>>> into hand-painted animation while preserving its motion",
"model": "kling-o1",
"mode": "std",
"duration": 5,
"generate_audio": false,
"video_list": [
{
"video_url": "https://example.com/source.mp4",
"refer_type": "base",
"keep_original_sound": "no"
}
]
}
```
Use `refer_type: "feature"` to reference style, motion, or a neighboring shot. Use `refer_type: "base"` (the default) to edit the supplied video. A base video cannot be combined with first/end frames.
#### Reference-video billing
1. Before submitting, check the current price for the model, mode and reference inputs. `kling-o1` and `kling-v3-omni` use a separate reference-video tier when `video_list` supplies a video; `image_list`, first frames and end frames alone do not trigger it. With a reference video, set `generate_audio=false`.
2. For successful reference-video requests, std costs **1.202 Credits per billed second** and pro costs **1.602 Credits per billed second**. For comparison, a 5-second `kling-o1` request costs:
| Mode | Without reference video | With 5 billed reference-video seconds |
|---|---:|---:|
| std | 4.005 Credits | 6.010 Credits |
| pro | 5.340 Credits | 8.010 Credits |
3. With `refer_type=base`, editing may follow the reference clip's length: final billing uses actual output duration rounded to integer seconds (5.708s bills as 6s; 5.041s as 5s). With `refer_type=feature`, billing uses the requested `duration`. Do not assume a base edit bills exactly the requested duration.
4. Poll the same task to completion and read the returned `usage.reference_video_seconds` for the final billed seconds. Reconcile the completed usage with the billing record; the submission estimate may differ from the final charge. Report only usage actually returned, not an invented response or cost inferred from the number of result URLs.
### 4. Extend Video
Continue an existing video with additional seconds.
```json
POST /kling/videos
{
"action": "extend",
"video_id": "existing-video-id",
"prompt": "the camera pulls back to reveal the full landscape",
"model": "kling-v2-5-turbo"
}
```
### 5. Motion Control
Apply precise camera/motion control from an image + reference video.
```json
POST /kling/motion
{
"image_url": "https://example.com/subject.jpg",
"video_url": "https://example.com/motion-reference.mp4",
"mode": "std",
"character_orientation": "image"
}
```
### 6. Lip Sync
Create a lip-synced video from a source video plus either an audio track or input text.
```json
POST /kling/lip-sync
{
"video_url": "https://example.com/source.mp4",
"mode": "audio2video",
"audio_url": "https://example.com/voiceover.mp3"
}
```
### 7. Talking Photo
Animate a still portrait from an image plus an audio track.
```json
POST /kling/talking-photo
{
"image_url": "https://example.com/portrait.jpg",
"audio_url": "https://example.com/voiceover.mp3",
"model": "kling-v2-1-master",
"duration": 5,
"mode": "pro"
}
```
## Parameters
| Parameter | Values | Description |
|-----------|--------|-------------|
| `action` | `"text2video"`, `"image2video"`, `"extend"` | Generation mode |
| `model` | See models table | Model to use |
| `prompt` | string | Required generation or continuation instructions |
| `mode` | `"std"`, `"pro"`, `"4k"` | Quality mode (`4k` only for `kling-v3` / `kling-v3-omni`, incompatible with `camera_control`) |
| `duration` | O1: `5`; v3/v3-omni/v3-turbo: integer `3`–`15`; others: `5`, `10` | Duration in seconds |
| `start_image_url` | URL | Required first frame for `action=image2video` |
| `end_image_url` | URL | Optional end frame for `image2video`; requires `start_image_url` |
| `video_id` | string | Existing Kling video ID required by `action=extend` |
| `generate_audio` | `true`, `false` | Optional audio for v3, v3-omni and v2-6 pro. Turbo always includes audio: omit this field or use `true` |
| `aspect_ratio` | `"16:9"`, `"9:16"`, `"1:1"` | Video aspect ratio |
| `cfg_scale` | 0–1 | Prompt relevance strength |
| `negative_prompt` | string | What to avoid in the video |
| `camera_control` | object | Camera movement parameters |
| `image_list` | array | Omni reference images for `kling-o1` / `kling-v3-omni`; each item has `image_url` and optional `type` (`first_frame` / `end_frame`). Up to 7 images without a reference video, or 4 with one, including first/end frames |
| `video_list` | array | One MP4/MOV Omni reference video for `kling-o1` / `kling-v3-omni` (3–10s, 720–2160px, 24–60fps, ≤200MB); item has `video_url`, `refer_type` (`feature` / `base`), and `keep_original_sound` (`yes` / `no`) |
| `callback_url` | string | Async callback URL |
| `mode` (`/kling/lip-sync`) | `"audio2video"`, `"text2video"` | Lip-sync mode |
| `video_url` (`/kling/lip-sync`) | URL | Source video URL for lip-sync |
| `video_id` (`/kling/lip-sync`) | string | Existing Kling video ID for lip-sync |
| `audio_url` (`/kling/lip-sync`) | URL | Audio source URL (for `audio2video`) |
| `audio_type` (`/kling/lip-sync`) | `"url"`, `"file"` | Audio input type (default `url`) |
| `audio_file` (`/kling/lip-sync`) | string | Audio file payload when `audio_type=file` |
| `text` (`/kling/lip-sync`) | string | Input text to synthesize speech (for `text2video`) |
| `voice_id` (`/kling/lip-sync`) | string | Voice preset ID used in `text2video` |
| `voice_language` (`/kling/lip-sync`) | `"zh"`, `"en"` | TTS language for `text2video` (default `zh`) |
| `voice_speed` (`/kling/lip-sync`) | number | TTS speaking speed (default `1.0`) |
| `mode` (`/kling/motion`) | `"std"`, `"pro"` | Motion generation quality mode |
| `character_orientation` (`/kling/motion`) | `"image"`, `"video"` | Character orientation source |
| `image_url` (`/kling/talking-photo`) | URL | Source portrait image |
| `audio_url` (`/kling/talking-photo`) | URL | Driving audio track |
| `model` (`/kling/talking-photo`) | `"kling-v1"`, `"kling-v1-6"`, `"kling-v2-master"`, `"kling-v2-1-master"`, `"kling-v2-5-turbo"`, `"kling-v2-6"` | Talking-photo model |
| `duration` (`/kling/talking-photo`) | `5`, `10` | Talking-photo duration |
| `mode` (`/kling/talking-photo`) | `"std"`, `"pro"` | Talking-photo quality mode |
## Gotchas
- `kling-o1` supports `duration=5` only; `kling-v3` and `kling-v3-omni` support flexible `3`–`15` seconds; most other models support `5` or `10`
- `mode=4k` is only available for `kling-v3` and `kling-v3-omni` and is incompatible with `camera_control`
- `generate_audio` enables optional synchronized audio for `kling-v3`, `kling-v3-omni`, and `kling-v2-6` pro. Turbo includes audio and rejects `false`.
- `end_image_url` is only for `image2video` action — it defines the last frame
- Omni references are supported only by `kling-o1` and `kling-v3-omni`; cite them as `<<<image_N>>>` / `<<<video_1>>>`
- Omni reference requests do not support `negative_prompt`, `cfg_scale`, `camera_control`, or `mode=4k`
- With `video_list`, `generate_audio` must be `false`; a base video cannot be combined with first/end frames
- `element_list` accepts existing platform-owned or verified preset IDs; raw external IDs are rejected. Standalone management is not published.
- Motion control (`/kling/motion`) is a separate endpoint from video generation
- Lip-sync is a separate endpoint (`/kling/lip-sync`) and requires `mode`; use `audio_url` for `audio2video` or `text` + voice fields for `text2video`
- Talking-photo is a separate endpoint (`/kling/talking-photo`) and requires both `image_url` and `audio_url`
- Read the current platform price for the selected model and quality mode. Their price ratios vary; Turbo standard/pro rates are 1.12/1.4 Credits per actual delivered second.
- Task states use `"succeed"` (not "succeeded") — check for this value when polling
- `negative_prompt` helps avoid unwanted elements (e.g., "blurry, low quality, text")
## Turbo, multishot and commerce
Turbo accepts text or a first frame. Omit `generate_audio` or set it to true; false is unsupported. It omits tail frames, camera_control, cfg_scale, separate negative_prompt and Omni references. Use `kling_generate_turbo_video` in the companion MCP or `kling turbo --request-file request.json` in the CLI.
V3/V3 Omni accept `multi_shot=true`. Automatic `shot_type="intelligence"` uses the global prompt. Customized shots use `multi_prompt` (1–6 entries) with consecutive index values, prompts of at most 512 characters and integer durations summing to the total. Companion tools: `kling_generate_storyboard` / `kling storyboard`.
Commercial capabilities use `/kling/goods-studio` and `/kling/video-commerce`. Pass structured contents/settings and poll task IDs through `/kling/tasks`. Apparel, virtual try-on and standalone asset management are not currently published.
## Reference inputs
Use `element_list` and `voice_list` only with already valid platform references and the supported model combinations in the current backend guide. Standalone asset creation and management are not published.
© acedatacloud · Apache-2.0 · AceDataCloud/Skills
Similar skills
All skills for AI and ML →kling-video by other authors comes first.