雀影

v1.0.1

雀影 - 你的 AI 视频创意伙伴,通过交互式引导帮你完善视频创意,生成 seedance、即梦提示词! 不同于简单的"描述→生成"模式,雀影会通过一系列精心设计的问题,帮助你: - 🎯 挖掘创意细节:从模糊想法到完整故事 - 🎨 推荐风格方案:根据内容智能匹配视觉风格 - 📝 优化台词音效:让角色更鲜活、...

0· 111·0 current·0 all-time
byfslong@fslong520
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description (AI video creative partner, generate Seedance prompts) matches the SKILL.md and the many included prompt/style guideline documents. The skill is instruction-only and requests no binaries, env vars, or credentials — which is proportionate for a prompt-generation / authoring helper.
Instruction Scope
Runtime instructions focus on eliciting user story, accepted media (images/videos/audio), producing time-stamped storyboards and Seedance 2.0–style prompts — all expected. Two items to watch: (1) the docs include many example prompts that assume or encourage photorealistic human faces, brand/product references (e.g., Coca‑Cola), and direct replication/replacement of素材 in source videos; platform limits are noted in docs (e.g., '不支持上传含有写实真人脸部的素材') but some examples contradict that restriction. (2) The skill expects users to upload up to a dozen media files and references @图片/@视频 tokens extensively — this is normal for a multimodal authoring tool but has privacy/copyright implications (the skill's instructions themselves do not sanitize or forbid PII or copyrighted inputs).
Install Mechanism
No install spec and no code files — the skill is instruction-only. That minimizes installation risk (nothing is downloaded or written by an installer).
Credentials
The skill declares no required env vars, no credentials, no config paths, and none are referenced in SKILL.md. The level of access requested is minimal and consistent with the described function.
Persistence & Privilege
always:false and no elevated privileges requested. The skill is user-invocable and may be invoked autonomously by the agent (default), which is normal. It does not request persistent system-wide changes or other skills' credentials.
Assessment
This skill is an instruction-only authoring tool for creating Seedance-style video prompts and storyboards. It's coherent and low-risk in that it asks for no credentials and installs nothing. Before installing or using it: (1) Be cautious about uploading personal photos or videos (especially real people's faces) — the docs refer to large multimodal uploads and some example prompts assume photorealistic faces even though the platform note says real faces may be disallowed. (2) Watch for copyrighted/trademark content in example prompts (brands, well-known characters); using or distributing outputs containing trademarks or copyrighted likenesses can have legal/usage limits. (3) Review any generated prompts before you paste them into external services (Seedance or image/video generators) to ensure they comply with platform rules and privacy expectations. If you need stronger assurance, ask the author for an explicit privacy/copyright policy or request a version of the skill that forbids references to real-person likenesses and commercial brands.

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

latestvk978wk12qhgsyspdzqwbypmcnx83n9jk

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments