Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

BigModel Image Video Generator

v1.0.0

使用 BigModel (CogView/CogVideoX) API 生成高质量图片和视频。当用户需要"生成图片"、"制作视频"、"AI 绘画"、"创建封面"、"设计海报"、"视觉内容生成"、或任何需要创建图像/视频内容的场景时使用此技能。即使没有明确提到"生成",只要用户需要创建、设计或制作视觉内容(如小说封...

0· 78·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for 156554395/bigmodel-image-video.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "BigModel Image Video Generator" (156554395/bigmodel-image-video) from ClawHub.
Skill page: https://clawhub.ai/156554395/bigmodel-image-video
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install bigmodel-image-video

ClawHub CLI

Package manager switcher

npx clawhub@latest install bigmodel-image-video
Security Scan
Capability signals
Requires sensitive credentials
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
high confidence
Purpose & Capability
Name/description match the implementation: the Python library and scripts call BigModel endpoints to generate images and videos. However the registry metadata claims 'Required env vars: none' while the SKILL.md, README, and lib/image_video.py clearly require BIGMODEL_API_KEY — a metadata omission/incoherence.
Instruction Scope
Run-time instructions are limited to setting BIGMODEL_API_KEY, installing 'requests', and running provided Python scripts. The SKILL.md and examples only reference the BigModel API domain (https://open.bigmodel.cn) and local files (lib/, scripts/). There are no instructions to read unrelated system files or exfiltrate data to alternate endpoints.
Install Mechanism
No formal install spec (instruction-only), which is lower risk, but the package includes Python modules and scripts that require 'requests' (README suggests 'pip install requests'). Because code files are present they will run locally and perform network requests — this is expected for the stated purpose but means code will be executed on the host.
!
Credentials
The skill requires one sensitive secret: BIGMODEL_API_KEY, but the registry metadata did not declare it (metadata lists none). The documentation recommends adding the key into ~/.zshrc (persists the secret in plaintext in a shell rc file), which may be inappropriate for some users. No other unexpected credentials or config paths are requested.
Persistence & Privilege
Skill does not request 'always: true' nor modify other skills or system-wide settings. It runs as a normal, user-invoked tool and only writes expected outputs (e.g., batch_results.json) to the current working directory.
What to consider before installing
This skill appears to do what it says (call BigModel to generate images/videos), but there are a few things to check before installing: 1) Metadata mismatch: the registry failed to declare BIGMODEL_API_KEY — treat the API key as required. 2) Secrets handling: the docs suggest appending the key to ~/.zshrc (persists it in plaintext). Consider setting the key only in the process environment or using a more secure secret store if the machine is shared. 3) Network calls: the code will send your prompts and payloads to https://open.bigmodel.cn/api — don't include sensitive or private data in prompts. 4) Dependencies & execution: the package runs local Python code and requires requests; review the included lib/image_video.py and scripts yourself and run in an isolated environment if unsure. 5) Recommendation: if you trust the source, update the registry metadata to declare BIGMODEL_API_KEY, avoid storing secrets in shared shell rc files, rotate the key after testing, and run the tool in a sandboxed environment. If you do not trust the author or cannot confirm the base_url/domain, do not provide your API key.

Like a lobster shell, security has layers — review code before you run it.

aivk978d807bx100553rhtxgbjw0x85bd2ncogvideoxvk978d807bx100553rhtxgbjw0x85bd2ncogviewvk978d807bx100553rhtxgbjw0x85bd2ngenerationvk978d807bx100553rhtxgbjw0x85bd2nimagevk978d807bx100553rhtxgbjw0x85bd2nlatestvk978d807bx100553rhtxgbjw0x85bd2nvideovk978d807bx100553rhtxgbjw0x85bd2n
78downloads
0stars
1versions
Updated 5d ago
v1.0.0
MIT-0

BigModel 生成图片和视频

使用智谱 AI 的 BigModel API 生成高质量图片和视频。支持单张/批量图片生成、视频生成(含 AI 音频)。


🚀 快速开始

1. 设置 API Key

必需步骤 - 只需设置一次:

# 临时设置(当前会话)
export BIGMODEL_API_KEY=your_api_key_here

# 永久设置(推荐)
echo 'export BIGMODEL_API_KEY=your_api_key_here' >> ~/.zshrc
source ~/.zshrc

获取 API Key:访问 智谱 AI BigModel 开放平台

2. 快速使用

最简单的方式 - 使用提供的脚本:

# 生成图片
python scripts/generate.py "一只可爱的橘猫"

# 生成视频
python scripts/generate.py "海边日落" --video

# 批量生成
python scripts/generate.py "日出 日落 星空 彩虹" --batch

# 查看所有选项
python scripts/generate.py --help

编程方式 - 在 Python 代码中使用:

import sys
sys.path.insert(0, '.claude/skills/image-video-generation/lib')

from image_video import generate_image

result = generate_image(prompt="描述内容")
print(result["data"][0]["url"])

📖 使用场景与示例

场景 1:小说/书籍封面

# 武侠小说封面(竖版)
python scripts/generate.py "中国武侠小说封面,水墨画风格,远山如黛,云雾缭绕,一把长剑插在岩石上,月光洒下" --size 1024x1792

# 言情小说封面
python scripts/generate.py "浪漫的粉色花瓣飘落,温柔的夕阳,温暖的色调" --size 1024x1792

# 科幻小说封面
python scripts/generate.py "未来城市,霓虹灯光,赛博朋克风格,高科技感" --size 1024x1792

场景 2:社交媒体内容

# 小红书风格图片
python scripts/generate.py "清新文艺风格,自然光线,极简构图,高饱和度"

# 朋友圈配图
python scripts/generate.py "生活记录,温馨日常,柔和光线"

# 头像生成
python scripts/generate.py "可爱的卡通风格猫咪头像,简洁背景"

场景 3:电商产品图

# 产品展示
python scripts/generate.py "白色背景上的蓝色运动鞋,专业产品摄影,柔和光线" --quality hd

# 场景展示
python scripts/generate.py "产品放在木质桌面上,温馨的家居环境,自然光"

场景 4:短视频制作

# 5秒短视频
python scripts/generate.py "一朵花在阳光下缓缓开放" --video --duration 5

# 10秒高质量视频
python scripts/generate.py "城市夜景,车流穿梭,灯光流动" --video --duration 10 --quality hd

场景 5:批量生成

# 批量生成不同风格的图片
python scripts/generate.py "日出时的山景 蓝色大海的海滩 秋天的森林小路 雪后的村庄" --batch

# 批量生成产品变体
python scripts/generate.py "红色款产品 蓝色款产品 黑色款产品 白色款产品" --batch

🎯 参数选择指南

图片模型选择

模型速度质量适用场景推荐指数
cogview-3-flash⚡⚡⚡⭐⭐⭐快速测试、大量生成、预览🌟🌟🌟
cogview-4-250304⚡⚡⭐⭐⭐⭐日常使用、平衡质量与速度🌟🌟🌟🌟
cogview-4⭐⭐⭐⭐⭐专业级、高质量输出🌟🌟🌟🌟🌟

建议:

  • 快速测试 → cogview-3-flash
  • 日常使用 → cogview-4-250304
  • 最终输出 → cogview-4 + --quality hd

视频模型选择

模型速度质量适用场景推荐指数
cogvideox-flash⚡⚡⚡⭐⭐⭐快速预览、测试效果🌟🌟🌟
cogvideox-2⚡⚡⭐⭐⭐⭐标准视频、日常使用🌟🌟🌟🌟
cogvideox-3⭐⭐⭐⭐⭐高质量长视频🌟🌟🌟🌟🌟

建议:

  • 快速测试 → cogvideox-flash
  • 日常使用 → cogvideox-2
  • 高质量输出 → cogvideox-3

尺寸选择

尺寸比例适用场景
1024x10241:1正方形、头像、社交媒体
1024x17929:16竖版、封面、海报
1792x102416:9横版、横幅、风景

💡 Prompt 编写技巧

好的 Prompt 特征

具体明确 - 描述主体、风格、场景 ✅ 细节丰富 - 包含光线、角度、氛围 ✅ 风格明确 - 指定艺术风格、质量要求

示例对比

❌ 不好:一只猫

✅ 好:一只橘色的短毛猫,坐在窗台上晒太阳,温暖的下午光线,柔和的景深效果,专业摄影

常用风格关键词

风格类:

  • 水墨画、油画、水彩、素描、卡通、写实
  • 赛博朋克、极简主义、复古、现代、传统

光线类:

  • 自然光、柔和光线、强烈对比、逆光、侧光

氛围类:

  • 温馨、神秘、浪漫、科技感、清新、厚重

质量类:

  • 专业摄影、高清、8K、电影级、精细细节

⚙️ 高级用法

直接使用 Python API

import sys
sys.path.insert(0, '.claude/skills/image-video-generation/lib')

from image_video import generate_image, batch_generate_images, generate_video, wait_for_video

# 单张图片
result = generate_image(
    prompt="描述内容",
    model="cogview-3-flash",
    quality="standard",
    size="1024x1024",
)
url = result["data"][0]["url"]

# 批量图片
prompts = ["描述1", "描述2", "描述3"]
results = batch_generate_images(prompts, max_concurrent=3)

# 视频生成
video = generate_video(prompt="描述内容", duration=5)
task_id = video["id"]
final = wait_for_video(task_id)
video_url = final["video_result"][0]["url"]

自定义并发控制

批量生成时控制并发数以优化性能:

results = batch_generate_images(
    prompts,
    max_concurrent=5  # 增加并发数(建议不超过 5)
)

📚 完整参考文档

详细的 API 文档和更多示例请参考:

  • 完整 API 参考:见 lib/image_video.py 中的函数文档
  • 示例代码:见 examples/ 目录
  • README:见 README.md 获取快速入门

❓ 故障排除

常见问题

Q: 提示 "需要设置 BIGMODEL_API_KEY 环境变量"

  • A: 未设置 API Key,参考"快速开始"第1步设置

Q: 生成速度很慢

  • A: 尝试使用 cogview-3-flashcogvideox-flash 模型

Q: 视频生成失败

  • A: 简化 prompt 或更换模型,检查网络连接

Q: 批量生成时中断

  • A: 降低并发数,使用 --concurrent 2

Q: 图片质量不够高

  • A: 使用 --quality hdcogview-4 模型

获取帮助

# 查看脚本帮助
python scripts/generate.py --help

# 测试 API Key 是否有效
python scripts/generate.py "test"  # 如果有效会生成测试图片

🎯 最佳实践

  1. 先快速测试 - 使用 flash 模型快速验证效果
  2. 再精细调整 - 满意后用高质量模型生成最终版本
  3. 保存好 Prompt - 记录有效的 prompt 供后续复用
  4. 批量处理 - 相似需求的图片使用批量生成
  5. 合理选择尺寸 - 根据用途选择合适的尺寸比例

Comments

Loading comments...