← Back to Skills Marketplace
TencentCloud Video AIGC Detection
by
Xinyu Wang
· GitHub ↗
· v1.0.9
· MIT-0
384
Downloads
0
Stars
0
Active Installs
8
Versions
Install in OpenClaw
/install tencentcloud-aigc-recog-video
Description
腾讯云 AI 生成视频识别 (TencentCloud Video AIGC Detection) 技能。适用于 AI 生成视频检测、视频真伪鉴别、AI 合成视频检测、Deepfake 视频检测等场景。TencentCloud Video AIGC Detection is an AI-generated con...
Usage Guidance
This skill appears to do what it claims: it calls Tencent Cloud's video moderation APIs and needs your Tencent Cloud SecretId/SecretKey and a BizType. Before installing or using it: (1) verify you trust the skill source (registry metadata omitted required env vars, so check provenance); (2) provide least-privilege credentials or use temporary STS tokens rather than long-lived root keys; (3) be aware that uploading local files will trigger a separate COS upload skill which may request additional credentials and installations; (4) confirm you are comfortable installing the public tencentcloud-sdk-python package and, if possible, run the scripts in an isolated environment or audit their network calls; (5) if you need stronger assurance, review the two included Python scripts (create_task.py and query_task.py) yourself — they implement straightforward Tencent Cloud SDK calls and are transparent about errors and required environment variables.
Capability Analysis
Type: OpenClaw Skill
Name: tencentcloud-aigc-recog-video
Version: 1.0.9
The skill bundle is a legitimate integration for Tencent Cloud's AI-generated video detection service. The Python scripts (create_task.py and query_task.py) use the official Tencent Cloud SDK to interact with the Video Moderation API, and the SKILL.md provides clear instructions for the agent, including safety rules to prevent the AI from hallucinating results. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found.
Capability Assessment
Purpose & Capability
The skill's name/description, SKILL.md, and the two Python scripts consistently implement calls to Tencent Cloud's VM CreateVideoModerationTask and DescribeTaskDetail APIs and therefore legitimately require Tencent Cloud API credentials and a BizType. Minor inconsistency: the registry metadata lists no required environment variables while the SKILL.md and scripts require TENCENTCLOUD_SECRET_ID, TENCENTCLOUD_SECRET_KEY and TENCENTCLOUD_AIGC_RECOG_VIDEO_BIZ_TYPE.
Instruction Scope
Runtime instructions stay within the stated purpose (create async moderation tasks, poll results). They require installing the Tencent Cloud SDK and — when a user provides a local file — instruct the agent to search for and install a separate Tencent COS upload skill and then upload the file to COS for a presigned URL. That behavior is coherent with supporting local-file uploads but means the agent will trigger additional installs and may require COS credentials.
Install Mechanism
This is an instruction-only skill with two small Python scripts; there is no automated install spec. The scripts require the public tencentcloud-sdk-python package (the scripts print a pip install instruction). No downloads from untrusted URLs or archive extraction are present.
Credentials
The only required secrets are Tencent Cloud SecretId/SecretKey (and optional STS token) plus a BizType identifier — these are proportional to calling Tencent Cloud's VM APIs. Note: local-file handling relies on a COS upload skill which may in turn require COS credentials; the registry metadata failing to declare the env vars is an administrative inconsistency you should be aware of.
Persistence & Privilege
The skill does not request always:true, does not modify other skills or system settings, and has no install-time persistence mechanism beyond standard use of environment variables. Autonomous invocation is allowed (platform default) but not excessive for this integration.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install tencentcloud-aigc-recog-video - After installation, invoke the skill by name or use
/tencentcloud-aigc-recog-video - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.9
TencentCloud Video AIGC Detection skill now provides a concise multilingual English/Chinese intro and improved cross-language instructions.
- Description and introduction sections updated for both Chinese and English audiences.
- Enhanced English explanations of core steps, options, and workflow.
- No code or behavior changes—usage and environment setup remain the same.
- Documentation streamlined by removing nonessential metadata and boilerplate.
v1.0.8
update SKILL.md
v1.0.7
tencentcloud-aigc-recog-video 1.0.7
- 文档描述优化：功能描述中新增覆盖 Veo3、Hunyuan、Jimeng、Hailuo 等主流模型生成的视频。
- Agent执行规范补充：“禁止跳过依赖安装”“禁止只打印脚本路径而不执行”等要求，进一步明确交互需返回真实结果。
- 其余使用方法、接口参数、环境变量等无变更，兼容原有用法。
v1.0.6
Version 1.0.6
- 增加了本地视频文件的处理说明：支持本地文件，说明需先上传至腾讯云 COS 并生成预签名 URL 后检测。
- 新增执行指令和流程说明中的“Step 0: 本地文件处理”环节，明确依赖 COS 上传相关技能。
- 在输入方式和限制说明中，更新为支持本地文件上传方式，并指明流程。
- 更新 metadata，声明需安装技能依赖、显式 requires 环境变量和 pip。
- 明确依赖由 Agent 手动安装，不再自动处理。
v1.0.3
No user-visible changes detected in this version.
- No file changes or documentation updates.
- Functionality and instructions remain unchanged.
v1.0.2
- Minor update with no user-facing changes.
- Documentation (SKILL.md) and core workflow remain unchanged.
- No impactful feature, bugfix, or behavior update in this version.
v1.0.1
- 更新环境变量名称为 TENCENTCLOUD_AIGC_RECOG_VIDEO_BIZ_TYPE，并统一相关用法与文档说明
- 完善环境配置指引，尤其是 BizType 获取和设置说明，确保用户能一次性完成所有配置步骤
- 明确首次环境检测出错时，需向用户完整展示全部配置指南
- 文档描述更简明，调整少量字段及表述，提升可读性和一致性
- 输出示例与错误处理指引同步细化，确保环境变量相关报错提示更加友好直观
v1.0.0
Initial release of the Tencent Cloud AI-Generated Video Recognition skill.
- Enables detection of AI-generated (AIGC) and deepfake videos via Tencent Cloud Video Moderation APIs.
- Supports asynchronous video moderation: creates a task and polls for results.
- Returns clear suggestions: "Pass" (real), "Review" (uncertain), "Block" (AI-generated).
- Provides detailed output including risky video snapshots and audio segments.
- Requires API credentials and BizType to be set as environment variables.
- Includes comprehensive setup instructions and strict agent operation rules to ensure accurate, API-based detection.
Metadata
Frequently Asked Questions
What is TencentCloud Video AIGC Detection?
腾讯云 AI 生成视频识别 (TencentCloud Video AIGC Detection) 技能。适用于 AI 生成视频检测、视频真伪鉴别、AI 合成视频检测、Deepfake 视频检测等场景。TencentCloud Video AIGC Detection is an AI-generated con... It is an AI Agent Skill for Claude Code / OpenClaw, with 384 downloads so far.
How do I install TencentCloud Video AIGC Detection?
Run "/install tencentcloud-aigc-recog-video" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is TencentCloud Video AIGC Detection free?
Yes, TencentCloud Video AIGC Detection is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does TencentCloud Video AIGC Detection support?
TencentCloud Video AIGC Detection is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created TencentCloud Video AIGC Detection?
It is built and maintained by Xinyu Wang (@shawn233); the current version is v1.0.9.
More Skills