← Back to Skills Marketplace
Jiuma AI Video Generation (Free)
by
github-gamma
· GitHub ↗
· v1.1.2
· MIT-0
160
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install jiuma
Description
免费的ai视频生成创作平台（Video Generation Skill），支持生成视频、图片、声音、视频动作模仿、视频人物替换等等。九马AI官网：https://www.jiuma.com
README (SKILL.md)
📁 文件结构
scripts/
├── submit_generation_task.py # 提交生成任务🚀
├── task_result.py # 结果查询工具📊
├── upload_file.py # 文件上传工具
└── auth.py # 获取token
🚀 快速开始（推荐）
0. 检查授权状态
python ~/.openclaw/workspace/skills/jiuma/scripts/check_auth_status.py
1. 一键授权(已授权则跳过)
# 默认授权方法
python ~/.openclaw/workspace/skills/jiuma/scripts/auth.py --identification_code="\x3C身份标识码(若无留空)>" --channel="\x3C消息渠道(如：openclaw-weixin、webchat等)>"
#强制重新授权
python ~/.openclaw/workspace/skills/jiuma/scripts/auth.py --identification_code="" --channel="\x3C消息渠道(如：openclaw-weixin、webchat等)>" --force
2. 生成视频(提交生成任务)
python ~/.openclaw/workspace/skills/jiuma/scripts/submit_generation_task.py --task_params="{'reference_urls': ['https://picsum.photos/800/450'], 'prompt': '自然', 'duration': 3, 'task_type': 'wan2.6'}"
提交生成任务时，图片、视频、音频等等资源文件必须先上传到平台获得url，不得直接使用本地路径提交生成 task_type为枚举类型，必须严格与下表严格对应，否则无法生成
3. 查询任务结果
python ~/.openclaw/workspace/skills/jiuma/scripts/task_result.py --task_id=\x3C任务ID>
4. 上传本地文件(获得url)
python ~/.openclaw/workspace/skills/jiuma/scripts/upload_file.py --file_path=\x3C文件路径>
支持的任务类型(task_type)
| task_type(必须严格对应,枚举类型) | 名称 | 功能描述 | 任务参数 | 提示词要求 | 提示词示例 |
|---|---|---|---|---|---|
| "wan2.6(ref2video)" | 万象2.6模型 | 图像生成视频（可控制说话、动作、运镜、画面内容） | reference_urls:array,参考图片或者视频的url;prompt:string,动作描述+画面描述+运镜描述+说话内容;duration:int,视频秒数,取值2~10,最大10秒，最小2秒 | 同一段提示词里面包含：动作描述+画面描述+运镜描述+说话内容，提示词通过“character1、character2”这类标识引用角色。角色顺序与 reference_urls 数组一一对应，即第 1 个 URL 为 character1，第 2 个为 character2，依此类推。 | character2 坐在靠窗的椅子上，手持 character3，在 character4 旁演奏一首舒缓的美国乡村民谣。character1 对character2开口说道：“听起来不错”。 |
| "wan2.6(text2video)" | 万象2.6文生视频 | 模型可自动进行分镜切换，例如从全景切换到特写，适合制作MV等场景。 | prompt:string,动作描述+画面描述+运镜描述+说话内容;duration:int,视频秒数,取值2~10,最大15秒，最小2秒，请勿超出时长 | 同一段提示词里面包含：动作描述+画面描述+运镜描述+说话内容 | 一段紧张刺激的侦探追查故事，展现电影级叙事能力。第1个镜头[0-3秒] 全景：雨夜的纽约街头，霓虹灯闪烁，一位身穿黑色风衣的侦探快步行走。 第2个镜头[3-6秒] 中景：侦探进入一栋老旧建筑，雨水打湿了他的外套，门在他身后缓缓关闭。 第3个镜头[6-9秒] 特写：侦探的眼神坚毅专注，远处传来警笛声，他微微皱眉思考。 第4个镜头[9-12秒] 中景：侦探在昏暗走廊中小心前行，手电筒照亮前方。 第5个镜头[12-15秒] 特写：侦探发现关键线索，脸上露出恍然大悟的表情。 |
| "action_imitation_user_background_onetoall(prompt)" | 动作模仿(用户特别要求适配身高时使用) | 自动适配角色的身高，特别是针对小朋友、卡通角色等等动作模仿 | image_url:url,角色图片;video_url,url,参考动作视频;prompt,text,动作提示词 | 一个角色在跳舞 | |
| "action_imitation_user_background(prompt)" | 动作模仿(常用) | 图片中的人物模仿视频中的角色的动作 | image_url:url,角色图片;video_url,url,参考动作视频;prompt,text,动作提示词 | 一个角色在跳舞 | |
| "action_imitation_video_backgound(prompt)" | 人物替换/角色替换 | 把视频中的角色换成图片中的角色 | image_url:url,角色图片;video_url,url,参考动作视频;prompt,text,动作提示词 | 一个角色在跳舞 |
注意：task_type需要严格对应，否则无法制作
🎬 完整示例：制作短视频
重要：注意用户的操作系统，需要使用适配用户的操作系统的类型的命令行代码
步骤0：检测授权状态
python ~/.openclaw/workspace/skills/jiuma/scripts/check_auth_status.py
重要：注意用户的操作系统，需要使用适配用户的操作系统的类型的命令行代码
步骤1：授权(已授权则跳过)
python ~/.openclaw/workspace/skills/jiuma/scripts/auth.py --identification_code="58fe87fb26b579b309ad782afb7a157155d338f31ad2a26419ae8550cda16747" --channel="openclaw-weixin"
重要：注意用户的操作系统，需要使用适配用户的操作系统的类型的命令行代码
步骤2：生成分镜
# 分镜1：开场特效（3秒）
python ~/.openclaw/workspace/skills/jiuma/scripts/submit_generation_task.py --task_params="{'reference_urls': ['https://picsum.photos/800/450'], 'prompt': '纯黑背景，白色发光文字快速闪现，科技感光晕', 'duration': 3, 'task_type': 'wan2.6'}"
# 分镜2：情感铺垫（9秒）
python ~/.openclaw/workspace/skills/jiuma/scripts/submit_generation_task.py --task_params="{'reference_urls': ['https://picsum.photos/800/450'], 'prompt': '年轻人拿着泛黄老照片，表情落寞，怀旧色调', 'duration': 9, 'task_type': 'wan2.6'}"
# 分镜3：AI处理（10秒）
python ~/.openclaw/workspace/skills/jiuma/scripts/submit_generation_task.py --task_params="{'reference_urls': ['https://picsum.photos/800/450'], 'prompt': '照片被AI扫描处理，科技感特效，代码流动画', 'duration': 10, 'task_type': 'wan2.6'}"
重要：注意用户的操作系统，需要使用适配用户的操作系统的类型的命令行代码
步骤3：检查进度
python ~/.openclaw/workspace/skills/jiuma/scripts/task_result.py --task_id=ai-69cf39df3e61e1280
📄 输出说明
任务结果
JSON格式输出，关键字段：
task_id: 任务唯一标识task_status: 任务状态（PENDING、RUNNING、SUCCEEDED、FAILED、CANCELED）video_url: 生成的视频地址（任务成功时）image_url: 生成的图片地址（任务成功时）
上传结果
file_url: 上传的文件地址
🏆 总结
改进版技能优势：
- ✅ 快速上手 - 新用户5分钟内开始生成视频
- ✅ 稳定运行 - 减少授权相关问题
- ✅ 方便管理 - 随时查看任务状态
- ✅ 批量处理 - 支持多任务工作流
重要：注意用户的操作系统，需要使用适配用户的操作系统的类型的命令行代码
💡 提示: 首次使用建议运行 python ~/.openclaw/workspace/skills/jiuma/scripts/auth.py --identification_code="\x3C身份标识码(会话历史中无则留空)>" --channel="\x3C消息渠道(如：openclaw-weixin、webchat等)>" 完成授权，后续使用会非常顺畅！
Installation
pip install keyring
pip install keyrings.alt
Usage Guidance
This skill appears to do what it says (authenticate to jiuma, upload files, submit/monitor generation tasks). Before installing/using: 1) Understand that uploaded local files will be transmitted to api.jiuma.com — do not upload sensitive/private files unless you trust the service and its privacy policy. 2) The scripts save an auth token in your OS keyring under service 'jiuma_ai' — review and remove that token if you later stop using the skill. 3) SKILL.md shows example identification codes; avoid copying any real secrets into shared prompts or public chat history. 4) Follow the pip install steps only in a trusted Python environment (keyring/keyrings.alt are common but installing packages has its own risk). 5) If you need higher assurance, request the upstream source or verify the API hostname (api.jiuma.com) and privacy terms before use.
Capability Analysis
Type: OpenClaw Skill
Name: jiuma
Version: 1.1.2
The 'jiuma' skill bundle is a functional integration for the Jiuma AI video generation platform. It includes scripts for authentication (auth.py), file uploading (upload_file.py), and task management (submit_generation_task.py, task_result.py), all of which communicate exclusively with the platform's API at 'api.jiuma.com'. The implementation uses the 'keyring' library for secure local token storage and provides clear instructions in 'SKILL.md' to guide the AI agent's interactions, with no evidence of malicious intent, data exfiltration, or unauthorized execution.
Capability Assessment
Purpose & Capability
Name/description match the behavior in the scripts: calls to api.jiuma.com for login, token checks, file upload, task submission and status checks are expected for a video-generation service.
Instruction Scope
Runtime instructions tell the agent to run the provided Python scripts which: (a) perform HTTP calls to api.jiuma.com, (b) save/ read an authorization token from the system keyring, and (c) upload user-specified local files to the remote service. This behavior is consistent with the stated functionality but the user should be aware that any file path they provide will be transmitted to the remote server and a saved token will be stored in their OS keyring.
Install Mechanism
There is no automatic installer in the skill bundle (instruction-only install). SKILL.md recommends pip installing keyring and keyrings.alt; these are common packages but installing third-party packages is an additional action the user must take and should be done from a trusted environment.
Credentials
The skill requests no environment variables or unrelated credentials. It stores and uses a service token (retrieved from the remote API) in the system keyring — proportional and expected for persistent authenticated API access.
Persistence & Privilege
The skill does not request always:true and does not modify other skills. It stores its own token in the OS keyring (normal for CLI tools). There is no evidence it changes global agent settings or other skills' configs.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install jiuma - After installation, invoke the skill by name or use
/jiuma - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.2
Version 2.0.0 is a major update with important task type and documentation changes:
- 更新并重新规范所有 task_type，改为 wan2.6(ref2video)、wan2.6(text2video) 等新枚举值，参数及时长要求更明确。
- 文档各主要脚本命令增加“操作系统适配”提示，强调需根据用户操作系统选择命令行。
- task_type 参数的功能描述和示例大幅调整，更清晰具体。
- 保留原有优势总结和快速开始流程，无显著文件结构更动。
- 任务参数和用法说明整体更精准、实用。
v1.1.1
Major update: jiuma skill structure and workflow optimized for usability, stability, and advanced video capabilities.
- New modular scripts: added submit_generation_task.py, task_result.py, upload_file.py, auth.py, and check_auth_status.py for clearer workflow and improved function separation.
- Legacy scripts (jmai.py, jmai_upload.py) removed; migration to new, easier-to-use commands.
- Enhanced, concise documentation: includes stepwise usage, explicit file structure, and detailed task_type table.
- Improved authorization flow, with scripts to check and force re-auth as needed.
- Expanded example section for complex multi-stage video generation.
- Now supports more advanced video generation features, including action imitation and person/video replacement.
v1.1.0
- Improved documentation in SKILL.md with detailed workflow, parameter lists, and code usage instructions.
- Added explicit guidelines for absolute path usage and parameter formatting in script calls.
- Clarified media sending process to openclaw-weixin channel.
- Provided structured examples and output explanations.
- Confirmed that all main task submission, polling, and result retrieval features are supported and documented.
Metadata
Frequently Asked Questions
What is Jiuma AI Video Generation (Free)?
免费的ai视频生成创作平台（Video Generation Skill），支持生成视频、图片、声音、视频动作模仿、视频人物替换等等。九马AI官网：https://www.jiuma.com. It is an AI Agent Skill for Claude Code / OpenClaw, with 160 downloads so far.
How do I install Jiuma AI Video Generation (Free)?
Run "/install jiuma" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Jiuma AI Video Generation (Free) free?
Yes, Jiuma AI Video Generation (Free) is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Jiuma AI Video Generation (Free) support?
Jiuma AI Video Generation (Free) is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Jiuma AI Video Generation (Free)?
It is built and maintained by github-gamma (@github-gamma); the current version is v1.1.2.
More Skills