NPAI SKILL
PassAudited by ClawScan on May 15, 2026.
Overview
This codeless skill appears to do what it says—create ecommerce video scripts—but users should notice that optional research or video generation may send selected product images and scripts to other tools.
This looks safe to install as an instruction-only script-writing skill. Before using the optional video-generation or batch-generation actions, make sure the product images are not confidential and that you are comfortable with any downstream video tool receiving the images, prompts, and scripts.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Product images, scripts, and related notes may be shared with a video-generation provider or another skill if the user chooses to generate videos.
This explicitly sends user-provided product context and image references to another tool. That is purpose-aligned for video generation, but it is a cross-tool data handoff users should understand.
If the user asks to generate the video after the script, pass the script and reference-image notes to the video generation skill/tool.
Only use the video-generation step for images and product details you are comfortable sending to the downstream video tool.
Batch video generation could consume time, quotas, or provider credits if invoked.
The skill can prepare or drive multiple downstream video-generation tasks. The artifacts include containment rules such as separate tasks and serial TVC execution, so this is purpose-aligned but still worth noticing.
Treat each script variant as a separate video task unless the downstream tool has verified multi-video output metadata.
Confirm before submitting batch or storyboard-first generation jobs, especially if the downstream video tool has costs or rate limits.
A 'directly generate video' action for TVC content may involve intermediate storyboard image generation before the final video.
The user-facing label may simplify a multi-step internal workflow that first creates storyboard images. This is not deceptive by itself, but it may understate the number of generation steps.
Do not expose backend workflow details in option names. ... If the chosen direction is商业广告/TVC, still write `直接生成视频`; the storyboard-first workflow belongs in the internal video handoff
If generation cost, timing, or intermediate outputs matter, ask the agent to explain the exact video-generation steps before proceeding.